A Factual Analysis of Analog vs. Digital… From a Digital-Lover’s Perspective
by ‘Miles Federman’
It was Jack White saying something
on the cable TV show ‘For What It’s Worth’ like, "Digital music is only about 20% of the actual sound that you would
hear from vinyl.... I don't think they (people) notice it as much with music,
how garbled it can sound," that set me off on my quest for real
answers.
Prior to that, Jack and his fellow vinyl worshipers had said things
like, “Digital doesn’t really sound any better than vinyl,” and, “We just love
the whole tangible experience of taking the record out of its jacket and
sleeve, placing it on the turntable and setting the needle down manually.”
The earlier claims, from about five
years ago, had a certain quaint charm that beckoned us all to step away from
the rat race for a few hours and just listen to some old records. Hell, even I
jumped onboard to a degree. But the more recent statements from that crowd have
sounded almost hateful toward a format that has brought unprecedented
portability and convenience to music lovers everywhere. And the claims made by
hipster digital-haters can be expected to escalate as long as they continue to
get props for it from all the other bearded steam punkers at the corner coffee
house.
I’m 40 years old, so I grew up
listening to my favorite music on vinyl records and cassette tapes. It wasn’t
until my junior year of high school that I finally broke down and bought a CD
player—five or six years after everyone else had. I was reluctant to replace a
lifetime of favorite albums on some johnny-come-lately format that would be
here today and gone tomorrow. However, after visiting a friend of my parents at
the age of 16 who worked as an executive at Columbia Records, I came away with
a new view of digital. He had a top-of-line Sony system (inventors of digital
sound technology) and the right pair of noise cancelling headphones. Going
through his CD catalog was like visiting the library of Congress, and I was
able to listen to any of them for as long as I wanted. Needless to say, I
returned home with a new respect for CD audio. The clarity, fidelity (similarity
to master recording) and lack of surface noise changed my view of the whole
situation. Needless to say, I’ve spent the last two and a half decades hunting
down all of my favorite obscure shit on digital. I’ve even taken analog tapes
of my old college band and transferred them to digital—all the while believing
that I’m giving the recordings a second chance at another life. I can’t just
sit still and let all of these Mumford wannabees drag us back to the
days of crackling vinyl and hissing tapes without at least doing some research
to either support or undermine my opinions!?!
So, I did a lot of internet
research. No, I didn’t cite references here, but rest assured every number and
fact has been cross-checked with more than two sources. Yes, this info was ALL
ripped off from other people who are probably the types of geeks who knew all
of this shit before they were twelve. If you find info in here that looks like
it was lifted from somewhere else…it definitely was.
It’s also important to know that I
wouldn’t have dug into this if I weren’t first and foremost a music lover. I
went into this thinking, “If vinyl really is better than digital, I owe it to
myself and my favorite artists to find out.” It does get a little sciencey in
places, but anything less in-depth would have been a waste of time. I did try
to keep it understandable.
So, here goes. A couple of
definitions:
Analog audio: an electrical signal that directly represents sound.
Digital audio: An indirect representation of sound by numbers.
Pulse Code Modulation (PCM) is the method used to digitally
represent analog signals. In a PCM
stream, a series of digital snapshots are taken of the analog sound wave at
equally spaced intervals, and each “snapshot” is quantized to the nearest value
within a digital framework. Since all speakers must be powered by an analog
signal, a digital-to-analog converter (DAC)
is employed upon playback. Two of the most important factors in determining the
quality and resolution of digital sound are the sample rate and bit depth.
Sample rate: (sometimes called the sampling frequency) is the
number of times an analog audio source is measured—or sampled—per second by a
digital recording device. Higher sample rates result in higher sound quality
because the analog waveform is more closely represented by the individual
samples. Common sample rates are;
44,100 samples per second (also known as 44.1 kHz audio) is commonly used for CDs and Mp3s. This
sample rate was set by Sony in the early 1980s after in-house research found
that digital audio sampled from analog at or above 40,000 shots per second is
perceived by most human ears as perfect sound. 44.1 kHz sample rate allows for approx. 80 minutes of two-channel
stereo music on one disc.
48,000 samples per second (aka 48
kHz) is used in almost all digital video formats as well as many
professional applications, and was originally adopted as the professional
alternative to the 44.1 kHz sample
rate which was intended more for consumer applications.
96,000 samples per second (aka 96
kHz) is a multiple of 48 kHz and
is the sample rate used for DVD audio.
It’s also becoming the professional standard for audio post-production and
music recording. If a home entertainment system is not equipped to handle 96 kHz audio, DVD players can convert the output to 48 kHz.
192,000 samples per second (aka 192 kHz) is also a multiple of 48
and 96 kHz. It is a very high-res
sample rate used mostly for professional music recording and mastering. Blu-ray audio as advocated by Neil
Young uses this sample rate. It is four times higher than the industry standard
for professional video editing equipment. Additional research by Sony has
indicated that sample rates above 192
kHz are impractical and tend to generate mysterious errors not seen in
lower sample rates.
When music produced at higher
sample rates is transferred to a CD
with 44.1 kHz audio, the process is
called down-sampling. Recording
industry professionals agree that CDs
sound better when the original recording and post-production work are done at
higher sample rates.
Bit Depth: describes the number of bits of information recorded for
each sample. Bit depth directly
corresponds to the resolution, volume and dynamic range of each sample in a
digital audio file. Because digital audio samples use binary numbers (ones and
zeros) to represent this dynamic range (the gap between the quietest sound and
the loudest), higher bit depth means
your audio sample is more accurately represented. 16-bit CD audio can represent 65,536 dynamic possibilities per
sample while 24-bit digital samples
can represent approx. 16,777,216. These different dynamic levels can be thought
of as rungs on a ladder. With 24-bit
sound, there are more rungs than with 16-bit
and the rungs are spaced closer together. When a digital recording device
measures the strength of a sample to fall in between two rungs, it simply
rounds to the nearest one. 16-bit
digital audio takes up less disc space and requires less computing power to
process while still containing enough subtlety and nuance to accommodate commercial
music applications. 24-bit digital audio
is necessary in cinematic applications where a movie character might whisper a
line in close proximity to a loud explosion going off. Normal DVD movies operate at 24-bit / 96 kHz audio.
MP3s (short for
MPEG Audio Layer-3) are compressed digital files that have no fixed bit depth. Their bit depth can vary depending on the particular frame of compressed
data. The decoding device (or MP3
player) determines the bit depth of
each PCM stream. MP3 uses a loss-inducing compression
algorithm to greatly reduce the amount of data required to represent the audio
recording…while still providing a faithful reproduction of the original
uncompressed audio. An MP3 file is
typically about 1/11th the size of the same CD file created from the original audio source. An MP3 file can also be constructed at
higher or lower bit rates* with
higher or lower quality resulting. The compression technology works by reducing
accuracy of certain parts of sound that are considered to be beyond the
auditory capabilities of most people. This method is commonly referred to as perceptual
coding. It uses modeling to discard or reduce background noise and frequencies
that are less audible or inaudible to human hearing. The MP3 software then records the remaining information in an efficient
manner. Initially released in 1995 as a way to download music files via dial-up
internet, MP3 has become the
standard for portable digital audio by combining small file size with good
sound quality.
*Bit
rate (or bitrate): refers to the number of bits—or the
amount of data—that are processed over a certain amount of time. In audio, this
usually means kilobits per second. For example, the music you buy on iTunes is
256 kilobits per second, meaning there are 256 kilobits of data stored in every
second of a song.
The higher the bitrate of a track, the more space it
will take up on your computer. Generally, an audio CD will actually take up quite a lot of space, so compressing files
to save room on a hard drive has become common.
Studio Master Tape: Many vinyl
records and CDs have been
created from quarter-inch reel-to-reel magnetic tapes which were originally
recorded at 15 inches per second. When a high quality analog tape is recording
at that speed, it offers approximately 80,000,000 individual particles per
second that can be magnetically stacked and aligned to store information. This
extremely high resolution outperforms all other formats when it comes to its
nearly unlimited sample rate and bit depth. Unfortunately, the unused particles
still make noise upon playback and this unnecessary sound is often referred to
as analog hiss. Digital filters are employed to minimize the noise during
digital re-mastering for CDs, DVD
and Blu-Ray audio.
Phonograph record technology was invented in 1877 by Thomas Edison
and originally used cylinders made of metal or wax. Emile Berliner, an American
inventor, demonstrated his Gramophone machine in 1888. This was the first
machine to make recordings on a flat, not cylindrical, disc. Phonograph record purists point
out that a record album should theoretically contain a perfectly uninterrupted reflection
of the original sound wave because sample
rate is not a factor in the vibrating stylus technology that writes the
groove. Science, however, tells us that vinyl records must have a sample rate. Only instead of being
based on snapshots per second, it’s based on the amount of vinyl passing under
the needle per second as well as other format constraints stemming from how
deep and/or wide a record groove can be. Unlike the constant sample rates
employed in digital recordings, sample
rates for vinyl records start
out high but are already diminishing rapidly from the moment the needle touches
down.
On a typical 12″ vinyl record, the ‘start’ groove, on the outside of the record
measures 5.75″ from the center. The ‘end’ groove measures at around 2.25″ from
center. Spinning at 33 revolutions per minute, the needle is traveling at
19.8 inches per second at the outside of the record, then gradually
slows to a meager 7.7 inches per second. That’s less than half the beginning
sample rate. Even at 45 rpm, the
starting speed is the best resolution available at a speedy 27.09 inches per
second, slowing to 10.59 inches per second in the center. Note that
a 45rpm record actually produces a higher-fidelity signal compared to a 33rpm,
since it’s moving 136% faster at the outermost point. This also demonstrates
the wider audio response available to vinyl spinning at 45rpm. The result is much
more bass. If we give vinyl records extra credit for their true analog
properties and say for sake of argument that their beginning sample rate is a
full one-third higher than CDs, a
few minutes into a 12” LP side, the
sample rate would have dropped below CD
quality in every case.
Vinyl has other limitations
inherent to the format. Surface noise results from the phonograph needle being
in contact with the record itself. Even brand new dust-free vinyl produces some
unwanted sounds.
The needle is also shaving off microscopic amounts of
vinyl—and therefore recorded information—each time a record is played. This
diminishes sound quality over time. When CDs
were introduced in the early 1980s, two major selling points were the lack of
surface noise and disc deterioration—both of which are made possible by a
data-collecting lens that never touches the CD itself. Another factor that limits the aural capabilities of
vinyl is the fact that the spiral groove on each side of the record cannot
touch or get too close to its neighboring groove, otherwise the record will be
unplayable. Since the record groove is packed to achieve two-channel stereo
(accomplished by a combination of up/down and side-to-side needle movements)
and a 24-minute maximum playtime per side, the sample size (groove width) is also lowered, offering less volume for the audio stream.
Since magnetic tape is now made by
only a few manufacturers, the cost has become nearly prohibitive for routine
use in recording studios. This is why more and more new studio master
recordings are 24-bit / 96 kHz digital.
Hence, a great many recent vinyl releases contain digital music that has been
converted to analog. Most vinyl enthusiasts consider that to be a big
no-no.
When digital audio was first
brought to market in the form of retail music CDs thirty years ago, most sound engineers and music industry
executives didn’t understand the unique production properties of the new
medium. This led to several older albums being rushed to market on CD without having been properly
re-mastered for digital. The result was a bunch of disappointing releases
of classic albums that had previously sounded good on vinyl. Fans and critics
alike called the CDs “tinny” or
“weak” and criticized the perceived inconsistencies of the new format. Legendary
producer Phil Spector was one of the first to point out that it wasn’t the new
digital medium, but an outdated mix on the old master tapes that had failed the
CD re-issues. More recent CD releases of these classic albums
have benefitted greatly from a re-mixed master which has been EQ’d specifically
for 16-bit digital audio.
Now, with all that being said, it
all comes down to YOUR ears. I can
honestly say that even though the evidence leads me to think that I was right
all along about digital being better, I couldn’t live with myself if I didn’t
admit that a couple of my favorite guitar solos do sound just a little bit more…well…shredding
on vinyl. And yes, that was hard for me to admit.
Lastly, a quick mind-blower for
you. Remember that number 65,536 from earlier? It’s the number of dynamic
possibilities in 16-bit digital audio. Well, since computer operating systems
approach numerous tasks in the same way, 65,536 was also the maximum number of vertical
boxes on a Microsoft Excel spreadsheet until 2007. Why? 65,536 is the number of
different values representable in a number of 16 binary digits, also known as a
short integer in many computer programming systems. That’s why a 16-bit
microprocessor chip can directly access 65,536 memory addresses and 16-bit
high-color computer graphics support a color palette of exactly 65,536
different colors. Since most people can picture what an Excel spreadsheet looks
like with its many boxes of information (including the ones that don’t fit on
your computer screen), one can imagine 16-bit
digital audio as presenting that much information to each of your two ears
at a rate of 44,100 times per second. For example, the spreadsheet below shows only 152 of the 65,000+ available cells.
