APJ Issue #9 - Vandersteen Audio

APJ Issue #9 - Vandersteen Audio
Auricle Publishing
Number 9 2003 $25.00 US
Source Components
Turntable Physics
Brooks Berdan
Blue Collar
Joe Harley
Disc Players
Ayre D-1x
Arcam FMJ DV27
Ayre CX-7
Opus 21
Verdi & Delius
Wadia 861
he Audio Perfectionist Journal has been published and distributed in electronic form since
1999. Journal #9 is the first one to be printed on
paper and distributed by mail and some background
information is called for.
The content in the Audio Perfectionist Journal is
based on my high fidelity approach to home music
reproduction. Journals #1 through #8 defined this
comprehensive method for evaluating audio components and systems. Goals were established, the
purpose of a home audio system was
described and we examined the functions of
some of the components that are vital to such
a system. The industry that manufactures and
sells audio components and systems was
scrutinized, and we challenged the
expertise of some of the writers
who review and criticize
the industry’s products.
Emotion is a Good Thing
We have established that our primary goal is the
emotional satisfaction that music can bring and we
need to think about how to assemble an audio system that will accomplish this goal. We are confronted
by a marketplace filled with a confusing myriad of
audio products which are often proffered with hype
and misinformation. Those who rely entirely on emotions to lead them to the components necessary to
bring about musical satisfaction in the home may
Journals #9 through #12
will delve deeper into our subject
of interest and provide information
about audio components that is
more specific and detailed.
You need this information if
you want to get the best value
for the money you invest in audio
components and the best sound
from the components you
Let’s start by restating our primary goal and outlining
some of the material that was introduced in previous
experience years of frustration and financial loss.
Wise enthusiasts will also enlist their intellect in order
to ferret out those products that are well designed
and fairly priced because the marketplace is filled
with products that are neither.
We seek an emotional connection with music but we must think
about the function of the components we choose in order to
assemble an audio system. We don’t want to waste money on
inadequate or poorly designed hardware and we do want to
fully exploit the capabilities of the good products we acquire.
the speakers and the listener within that room. Journal #3
begins the discussion of system components that continues
through Journal #8.
The Natural Order of Things
Reason is a Good Thing, Too
The Audio Perfectionist Journal is a publication for those
who respond emotionally to music and want to learn how to
choose audio components which allow that response. There
are many ways to go wrong and lots of advertising dollars
available to lead you astray.
Music has life-enriching value. Some audio components convey
the musical message better than others. Figuring out which
“Our primary goal is the
emotional satisfaction that
music can bring.”
components are best requires technical examination and experienced listening and the Journal can help you learn how to do
both. We evaluate the quality of audio components by listening
to them but we also understand how audio components work
and we know how to gauge the competence of the designer.
Knowledge makes it harder to fool us with spectacular but artificial sound or with impressive but false rhetoric. Remember,
misinformation is just like real information but wrong.
A little knowledge can work for you too. Wouldn’t you like to be
harder to fool? You have to make the ultimate decision about
which products to choose. The Journal will provide useful
information that will help you avoid costly mistakes.
The article The Natural Order of Things in Audio Perfectionist
Journal #3 follows the signal path through a home audio system and divides the system into sections based on component
function. There are three major component groups, or system
sections: source components, which provide information
retrieval; amplification components, which raise the signal
level; and output components, which convert that signal into
sound. Cables are an additional component group that will be
considered separately.
Each of the Audio Perfectionist Journals #9 through #12 will
concentrate on a system section, or component group. W e’ll
examine each section of an audio system in greater depth
while discussing the components that provide the specific functions of information retrieval, amplification and reproduction.
This journey will begin with Journal #9 which is all about
source components.
Source components are often undervalued by novices but the
source component is vitally important because it sets the limits
of resolution for the entire system.
Examining an Audio System, Section by Section
Following the signal path and grouping components based on
function makes it easier to understand what each component
does and how each individual component relates to other components and to the system as a whole. Understanding these
concepts is important because it helps us to properly allocate
resources in order to assemble a balanced audio system that
provides maximum performance for investment. A lack of
understanding can lead to frustration and financial waste.
The Basics
Journal #1, available free on the Audio Perfectionist website
(www.audioperfectionist.com), explains the philosophy of the
high fidelity approach to home music reproduction and home
theater. Some common myths are debunked and the basic
goals of the Journal are established. Journal #2, which is also
available free on the website, explains how the environment
affects the sound you hear and suggests the best ways to
improve listening room acoustics and the best ways to position
If you believe that speakers can reproduce a signal that doesn’t
come out of the amplifier, or that the amplifier can amplify a
signal that didn’t come out of the preamplifier, you will have a
difficult time properly allocating your resources while assembling an audio system that can provide satisfying sound. If you
forget that the source component establishes the limits of resolution for the entire system you may spend years and untold
sums of money futilely trying to reproduce a signal that was
Analog and Digital
never retrieved from the recording.
Many people spend too much on output components and too little
on amplification and source components. They believe that
speakers make the biggest difference in sound so they spend
most of their budget on speakers and shortchange the rest of the
system. The result is likely to be unsatisfactory at best. It is common to see systems that include $10,000 speakers reproducing
sound sourced from a $600 CD player. It is common and it’s stupid.
Factual, logical information can help you avoid costly and frustrating mistakes when choosing components for an audio system or deciding which component to upgrade first in the system
you already own. Unfortunately, hype and ridiculous claims are
often substituted for facts and logic. We’ll try to alert you to
claims that are over-the-top or just plain false and present logical arguments instead.
Let’s begin with some simple, irrefutable facts.
Truth and Logic
Here are some inarguable truths and some simple logic:
Speakers reproduce the signal that is delivered by the amplifier. The amplifier amplifies the signal that is delivered by the
preamplifier. The preamplifier processes the signal that is delivered by the source components.
No system component can create a musical signal. Each component is limited to reproducing the signal that is presented at
the input to that component. The source component retrieves
the signal from the recording medium, establishing the limit of
resolution for the entire system.
Today our audio systems can play music recorded on analog
media like magnetic tape and vinyl records, or music that has
been digitally encoded and stored on optical discs. This
“The source component
retrieves the signal from the
recording medium, establishing
the limit of resolution for the
entire system.”
Journal will provide basic information about analog and digital
signals and the components that are used to retrieve those signals from the recording. We will offer suggestions, based on
our experience with many components, for achieving the best
possible performance from analog and digital sources.
We, of course, haven’t heard every product available and information will be provided to help you judge components that
aren’t mentioned. Products that aren’t specifically mentioned in
this Journal are not necessarily bad (or good). We try to
refrain from commenting on components with which we have
little personal experience but that doesn’t mean that you
shouldn’t investigate them. A lot of stuff is available for your
consideration and we can evaluate only a small percentage of
The Current State of Digital
If an optical disc player or turntable system doesn’t retrieve the
signal from the disc or record, it’s lost forever. The preamp
can’t process a signal that doesn’t exist, the amplifier can’t
amplify that signal, and speakers—no matter how expensive or
exotic—can’t reproduce that signal. If you want high quality
sound to come out of your audio system you must put a high
quality signal in.
Of course each component can add undesirable artifacts like
noise and distortion to the desired signal, as described in
Journal #3, and those undesirable additions must be minimized for satisfactory results.
Digital audio has evolved and new formats like SACD and high
resolution LPCM (available on some DVD-Video and DVDAudio discs) provide hope for even better sound. That doesn’t
change the fact that the audio industry has languished for over
20 years while we have tried to create a silk purse from the
sow’s ear that is the compact disc. Today’s best 16-bit/44kHz
recordings played on today’s best CD players provide sound
that is hard to fault in hi-fi terms. Analog records and SACD
discs offer more.
Analog records and SACDs make it easier to become emotionally connected to the music. This emotional connection is the
hook that perpetuates the high-end audio industry and perhaps
the music business as well. Both are in a slump and I can’t
help but think that the compact disc is at least partly responsible.
Today’s best CD players are barely adequate as a source for a
high resolution audio system and decent CD players are still
quite expensive. An inexpensive turntable or SACD player will
convey the musical message far better than the best CD players available. Hi-fi buffs may never understand this but listeners seeking an emotional response from music will “get it” with
their first exposure to good analog or SACD.
I’m not going to tell you to throw
away your CD collection
and I’m not going to part
with mine. But you
need to know that
you are working
harder to achieve emotional satisfaction if compact disc is your primary source
for music. The contents of this
Journal will present a strong case for that position in the hope
that people will not simply settle for the mediocrity of CD and
will seek superior formats which are coming to market now
and, in the case of analog, have always been available.
This Journal won’t be devoted exclusively to criticizing the
compact disc. We will tell you how to get the best sound possible from your CD collection. We’ll review an array of players
and tell you which ones are worthy of your consideration and
which ones aren’t. We’ll compare the sound from CDs to the
sound from high resolution digital formats like SACD and high
res LPCM and we’ll compare players compatible with these formats.
The Current State of Analog
The general public may think that vinyl records became obsolete in the early 1980s when the compact disc arrived. This
point would be hard to argue based on sales numbers. Analog
records have never completely vanished from the market but
they represent a small niche market much like high-end audio.
These niches are of interest to only a small number of discerning music lovers. The fact that records won’t die demonstrates
a lasting appeal that can only be attributed to sound quality.
Vinyl records are a pain in the butt. They are easily damaged
and need to be cleaned and carefully stored. Some surface
noise is audible on most pressings and a high percentage of
new records are imperfect in some way.
You can’t listen to a record while riding in your car or jogging.
Records must be played on delicate mechanical devices that
are microphonic and easily damaged during house cleaning.
These devices (turntables) must be carefully adjusted after
installation and meticulously maintained. What a hassle.
Why would people put up with this “primitive” format in a digital
age? Because records sound better. The payoff for all the trouble is sound that is alive and deeply satisfying. Give records a
try and see if you don’t agree.
Source Components
This Journal is all about source components. We’ll start with
analog and finish with digital. Before we begin our examination
of source components I’d like to acknowledge some contributors who helped make this Journal possible.
New Members on the Journal Team
Shane Buettner has joined the Audio Perfectionist Journal as
Equipment Review Editor and he has contributed a large volume of work, including many product reviews, to this issue.
Shane is an experienced listener and journalist whose articles
and reviews have been published in Widescreen Review, The
Perfect Vision, and The Absolute Sound magazines. Shane and
I share a common vision about how audio publications should
serve consumers and we have been working closely together
for several years.
Rick Johnson has joined the Journal as Art Director. Rick’s
artistic skills will make this a more attractive and professional
publication that is better able to accurately convey information.
We’ll now be able to print photographs of products and people
along with drawings that will help to clarify complex subjects
that are difficult to describe with words alone.
I’d also like to express appreciation to the other contributors to
Journal #9: Joe Harley of Harley Music Productions, Brooks
Berdan of Brooks Berdan Ltd., and Chris Fitzgerald, a Journal
subscriber. Edith Hardesty has edited the thousands of words
of copy to correct our spelling and grammatical errors, as she
has done since the beginning. APJ
An analog signal has continuous physical variables as
opposed to a digital signal
which is composed of sequential samples that represent
numerical quantities at an instant
in time. An analog recording of
music contains a continuously variable signal that is directly analogous to the air pressure
variations, or “sound waves,” at the original event. A digital
recording of music contains a digital code consisting of
sequential samples that represent the amplitude of the analog
signal at specific instances in time or the change in amplitude
between one sample and the next.
Recordings are created by converting mechanical energy to
electrical energy and storing a representation of that electrical
energy. During playback electrical energy is converted into
mechanical energy to produce sound.
Energy conversion is performed by electromechanical transducers, including microphones, cutting heads and loudspeakers. Several conversions may take place—for vinyl records,
sound pressure is converted to an electrical signal at the microphone; the electrical signal is converted to mechanical energy
at the cutting head; mechanical energy from the record groove
is converted to an electrical signal in the pickup cartridge; and
electrical energy is converted to mechanical energy by the
loudspeakers. While electromechanical conversion occurs during recording and playback, the signal on an analog recording
is always a direct analog of the sound pressure and is continuously variable rather than sampled.
Tape and Vinyl
Analog sources available for high-end audio enthusiasts
include analog tape and vinyl records. Both are becoming
increasingly less common. Analog tape decks for home use are
almost extinct and we’re not going to spend any time discussing them. We won’t ignore analog tape completely
because many high quality recordings, even those that end up
on a digital medium like CD, are initially captured on analog tape.
Vinyl records, which have traditionally been our highest resolution source, are making a strong comeback in our niche market
and we will devote a substantial portion of this Journal to an
examination of analog turntable systems so you’ll know how
they work and how to choose a good one should you decide to
see what all the fuss is about. I hope that you do decide to
investigate records.
Analog Tape
An analog
energy in
a varying
field created by
magnetizing ferrous (containing iron) particles which are bonded to mylar tape. These particles pass by a recording head
which magnetizes them using a high-frequency bias current
along with the electrical signal that represents the musical
information captured by the microphone. During playback an
electrical signal is produced in the playback head by the passing magnetic field on the tape. This electrical signal is amplified
for conversion back to mechanical energy by the loudspeaker.
A tape recorder stores information magnetically utilizing record
and playback heads containing coils of wire which become
electromagnets for recording and generators for playback.
The record head becomes a variable electromagnet when the
signal and bias current flow through its coil. The playback head
becomes a generator when its coil is exposed to the variable
magnetic field from the ferrous (iron) particles on the recorded
An analog tape recording stores energy magnetically. Energy is
stored on vinyl records by physically carving it into the groove
walls of a master disc and transferring those groove wiggles to
a vinyl record.
Vinyl Records
Vinyl records store the recorded signal in the wiggles of the
grooves. The inner groove wall contains the signal from the left
and the
the signal from
the right
channel. The groove walls have a 90 degree included angle
and the stylus moves at ±45 degrees from vertical in response
to a signal that occurs only in one channel or the other. If equal
amplitude in-phase signals are recorded in both channels the
playback stylus will move laterally. Equal amplitude out-ofphase signals will produce vertical stylus movement
A cutting lathe is used to make records which can be created
from analog or digital tape recordings or directly from the
microphone feed—“direct to disc.” The cutter head converts the
electrical signal into mechanical motion and carves a groove in
When vinyl records are made the signal is equalized during
recording and playback to limit groove spacing and to minimize
surface noise. Low frequencies are reduced in amplitude and
high frequencies are increased in amplitude during recording
by applying the RIAA equalization curve. An inverse EQ curve
is applied during playback where bass frequencies are raised
in level and high frequencies are reduced in level to return the
signal to flat response.
Vinyl records can have a useful bandwidth of 10Hz-30kHz with
excellent phase response. Musical dynamic range is audibly
superior to CD even though signal-to-noise ratio numbers might
lead slide rule jockeys to believe otherwise. A playback stylus
in intimate contact with the groove walls can retrieve recorded
information that is remarkably subtle. Listen and compare CD
and analog versions of the same recording and hear just how
much micro detail—most notably instrumental decay and high
frequency “air”—is missing from the CD yet clearly audible on
the record. Or attend an Audio Perfectionist seminar and I’ll
demonstrate this for you.
The Analog Difference
If you find it more difficult now to respond to music with the
degree of enthusiasm you had in your youth, or if you are a
youth and never completely understood those who talk of a
deep emotional connection to music, try listening to vinyl
The sound from records is so
much more “alive”
and involving.
the lacquer coating on the surface of the aluminum master
disc. A vinyl record is produced from this master disc (several
intermediate stages are involved). During playback, the pickup
cartridge converts the groove wiggles into an electrical signal
that can be amplified and reproduced as sound.
The process is tedious but uncomplicated and the music
remains intact all along the way. The master disc and the stampers which impress the signal on the vinyl discs can be played.
(Stampers require a special stylus because a stamper is like a
photographic negative—grooves become ridges.)
records. They don’t have to be new audiophile recordings. In
fact, it’s probably better if they aren’t. Open yourself to this
experience and you may find something that’s been missing in
your life. Let me tell you about what happened to me.
When I advanced into middle age and left the retail audio business, I got lazy. My turntable (a Linn LP-12 circa 1980) needed
a new belt and I had a state-of-the–art CD player (a W adia
860, now an 861). I had duplicated many of my favorite records
with CD copies of the same recordings. The CD sound was
impressive and everything was sooo easy. No disc cleaning, no
stylus cleaning, no hassle. And I could play CDs in my car.
Even though
it was now
more convenient to select
and play a
disc, I found
that I was listening to my
audio system
less and less.
When I
played music
that, in years past, had produced a deep emotional response,
the sound seemed spectacular and hard to fault (in hi-fi terms)
but the goose bumps no longer appeared. I decided that I just
had to accept a less thrilling and less satisfying involvement
with music—just as I had accepted other diminutions of excitement and joy in my life as a “natural” part of growing older.
CDs were supposed to provide “perfect sound forever.” I had
listened to dozens of CD players and I knew that I had one of
best available.
demonstrated this phenomenon for hundreds of people with
universal success.) Records make it easier to connect with the
music and the musicians.
Emotions restored
Except for the reintroduction of the turntable, all components in
my system remained unchanged. Many of the records that I
had chosen for this first listening session were the analog versions of the same recordings that had been emotionally unsatisfying on CD. The excitement wasn’t gone—it had simply been
stifled by inadequate digital coding.
I spent the decade of the ‘80s, while I was still selling audio
systems, trying to convince people that the CD was a step
backwards from the vinyl record, but in the ‘90s I had given in
to the convenience of what, by then, had become a much
improved digital recording and playback system. CDs are better
now since the dark ages of the 1980s but they’re still not good
After returning to my turntable as a source I have been reborn
as an analog druid. Although I have an outstanding CD-based
audio system, records simply sound better. Vinyl delivers the
musical goods like CD never could.
The Road Back to Analog
When I started to review audio products and write about audio
subjects I needed a complete reference system. I contacted my
old friend Ivor Tiefenbrun at Linn in Scotland and asked for the
parts to bring my antique LP-12 up to date. I upgraded the
turntable to Cirkus status and added a Lingo power supply and
the latest Ekos tonearm and Arkiv cartridge. When the turntable
system was assembled and operational, I sat down for what
turned out to be an ear-opening experience.
I played a bunch of 20-year-old records that used to “get me
off” in the old days and tears came to my eyes. The thrill wasn’t
gone, after all, or even diminished. In fact, because age and
experience have made me less self-conscious and more selfassured, I could really let myself go and I luxuriated in the
music like a pig in a mud bath.
The sound from records is so much more “alive” and involving.
The rhythm and pace of the music are easier to follow and
respond to. (I have no technical explanation for how CDs can
have a negative impact on rhythmic integration but I have
This was not a momentary experience and I continue to choose
analog as my preferred source whenever possible. Records
deliver music that is more rhythmically involving and more
emotionally gratifying. The message of the composer and the
skill of the artist are far easier to perceive when the source is
vinyl. I have a lot of compact discs and I continue to listen to
them but I buy CDs only when a vinyl or SACD version of a
recording is not available. APJ
Records Need Record Players
To play vinyl records you need a turntable, tone arm
and cartridge (along with a phono preamp—to be
covered in Journal #10). Before choosing these components you should know a little about the physics
involved. The following articles will provide necessary information.
Turntable Physics
If you want to play records you
need a record player.
Record players are
called turntables in
high-end audio language
but a high-end “turntable” is actually a system with three major components: the turntable
mechanism, which rotates the record; the tonearm, which
caries the pickup cartridge over the record surface; and the
pickup cartridge, which generates electrical output.
The job of the turntable system seems deceptively simple but
there is more to it than meets the eye. A high-end turntable
system must retrieve signals from the record grooves that can’t
be seen without the aid of a microscope. It must retrieve these
tiny signals while both resisting the powerful energy that bombards the system from the outside and dissipating disruptive
energy created within the turntable system.
Disruptive energy from within the system includes vibration
from the motor and main bearing and excitation of the tonearm
tube by the stylus motion in the groove. Disruptive energy from
outside the system includes footfalls and other mechanical disturbances, and acoustic energy from the loudspeakers.
Acoustic energy from the speakers may enter the turntable system from below through the support furniture or directly through
the air into the turntable components, particularly the light
weight tonearm tube.
The turntable motor is the only active component in the system.
The turntable platter, which supports the record, spins freely
on the main bearing and is driven by the motor. The quality
of the bearing which supports the platter is critical to performance. Tonearms
and cartridges are passive
components dependent on
the turntable motor for function. The motor drives the
platter which drives the
groove. The groove drives
the tonearm and the stylus.
record in three dimensions and act as a sump for energy from
sources including the speakers and the vibration from the stylus that is tracing the grooves.
The cartridge generates output when there is relative movement between the cartridge body and the stylus. Ideally the
cartridge body will be held motionless above the record while
the stylus tracks the groove, but if the stylus is motionless and
the cartridge body is moved by external energy which enters
the system, or by arm tube resonance reflected back to the
cartridge body, a signal will be generated which is unrelated to
the recorded information.
We’ll discuss the three system components in order of their significance to sound. First the turntable, then the tonearm and
finally the cartridge. While this order of sonic significance may
seem counterintuitive, experience will prove its validity.
The Turntable
The turntable has a motor which powers the entire system, a
rotating platter that supports the record, a main bearing assembly that supports the rotating platter, and a chassis that supports the main bearing and the tonearm and couples the two
together over a wide range of frequencies (ideally). This coupling is vitally important because it minimizes the effects of
vibration on the retrieval of information. Here’s how.
If the turntable subchassis and tonearm assembly provide rigid
coupling between the main bearing and the pickup cartridge,
noise and vibration entering the system will not cause relative
The tonearm must position the
cartridge accurately above the
Turntable Physics
movement between the stylus and cartridge body because that
noise or vibration will appear everywhere within the coupled
system, in phase.
If the cartridge body is decoupled from the main bearing any
noise or vibration that enters the system may cause relative
movement between the stylus and cartridge body which will
affect the retrieval of information or create a signal that is unrelated to the recording. Bearing noise would appear at the platter and also at the cartridge body in an ideal coupled system
and produce no output. Bearing noise would only appear at the
platter in a decoupled system and would be interpreted as a
signal by the cartridge.
Decoupling can occur at the main bearing-to-chassis junction,
at the tonearm-to-chassis junction, at the tonearm support-totonearm junction, at the tonearm-to-cartridge junction, etc.
Perfect coupling at all frequencies cannot be achieved in the
real world. Energy that cannot be coupled in phase must be
absorbed and dissipated rather than reflected. Mass isolated
turntables attempt to direct energy into massive components
that act as sumps in order to prevent this energy from being
reflected back into the system.
drive method for high-end audio but not the only one.
Motors turn with sequential pulses which are smoothed by the
flywheel action of the platter and the elasticity of the belt. As
the flywheel gets heavier the motor power must be increased.
The more powerful the motor the stronger the pulses. A motor
with just enough power to start the platter is usually the best
choice for smoothest rotation.
In general, DC (direct current) motors are voltage controlled
devices and AC (alternating current) motors are frequency controlled devices. DC motors facilitate continuously variable
speed contol, which is useful for turntables used by disc jockeys. AC motors are as speed accurate as the frequency of the
alternating current that drives them and don’t need servos or
sensing devices. DC motors use servo speed control systems
in virtually all commercial turntable designs. Servo-controlled
systems are, by definition, always turning at the wrong speed.
Servos can actually create flutter distortion at the sample rate
of the servo. Heavier platters smooth the servo action into a
more stable rotation. Heavier platters require more powerful
motors which create more powerful pulses.
The best turntables I’ve heard use AC motors and belt drive
systems and have platters that are moderately but not excessively heavy.
The Motor
An electric motor rotates the platter. The motor can be directly
attached to the platter—direct drive—or a detached motor can
rotate the platter using a belt or string drive. Motors rotate with
sequential pulses (in small jerky motions) as poles are energized. Fewer poles mean fewer, stronger pulses (jerks) per revolution. More poles produce more jerks
with less energy
per jerk.
A direct drive
motor must turn
at platter speed.
A belt drive motor will turn at a much higher speed
while rotating a small pulley that will drive the larger platter through the elastic belt. The belt can isolate the small jerky motions of the motor from the
platter which acts as a flywheel to smooth rotation
and stabilize speed. Belt drive is the preferred
The Platter
The turntable platter must support the record and rotate at a
constant and correct speed. Slow variations from the correct
speed are called “wow”
and rapid variations
are called “flutter.”
Wow and flutter
are descriptive
names that accurately convey the
negative sonic
impact of these distortions. Both
affect instrumental pitch and harmonic relationships.
Constant but incorrect speed also
affects pitch and harmonic integra-
Turntable Physics
tion. The turntable must rotate at the correct speed to accurately reproduce music. Speed can be verified with a strobe disc or
by playing a test tone of known frequency and using a frequency counter or oscilloscope to verify the frequency that is reproduced.
A heavier platter will provide more flywheel effect than a lighter
platter but the heavier platter will create far stronger forces in
the main bearing and will require a more powerful motor that
will introduce stronger pulses (jerks). Heavier is better up to a
Clear plastic platters look cool but provide no other benefit that
I can determine and they may increase static electric charge
because they are non-conductive. A metal platter can be electrically grounded to minimize static build-up in the record. A
metal platter can be electrically grounded to the tonearm.
The Main Bearing
The turntable platter rotates on a main bearing that must prevent wobble and create little noise. Low frequency noise, which
is created primarily in the platter bearing, is called “rumble.”
Rumble is a very destructive type of noise because it occurs at
low frequencies which can use up large amounts of amplifier
power and speaker excursion, producing intermodulation distortion and reducing headroom and dynamic range.
Every imaginable bearing type has been tried and the vast
majority of commercial turntables use a simple sleeve bearing
assembly with a thrust plate, which has proven to be the best
and longest-lasting solution.
Oil, which is essentially non-compressible, can fill the gap
between the spindle and the sleeve bearing providing lubrication and rigid coupling. An oil-bath bearing can be virtually free
of wobble. Air bearings rely on air, which is a compressible
substance, to support the platter and reduce friction between
the spindle and the sleeve and thrust plate. Air bearings
increase complexity and reduce coupling while providing no
audible benefit in my opinion.
The best turntables I’ve heard use oil-bath sleeve bearings.
Inexpensive turntables may effectively use dry bearings that
will suffice for several years.
Two basic methods are used to isolate turntables from external
energy and minimize acoustic feedback. One method utilizes a
spring suspension system that acts as a low-pass filter blocking
vibration at frequencies above the resonant frequency of the
suspension. The other method relies on mass. Because little
engineering is necessary, mass isolation is more popular with
small manufacturers.
Spring suspension systems are generally tuned to resonate at
3-4Hz. These suspensions are very effective at blocking vibration above 4Hz or so from entering the turntable system
through the supporting furniture below. Because they act as
low-pass filters they are ineffective at frequencies below 3-4Hz
and they make the turntable system very sensitive to footfall
disturbance. In a home with wood floors a spring suspended
turntable should be supported by a shelf mounted on a loadbearing wall. A suspended turntable should not be placed on an
isolation platform that includes another compliant system. T wo
compliant systems may interact with unpredictable results.
Spring suspension can’t isolate the turntable from airborne
energy which enters the turntable system directly rather than
through the supporting furniture. Rigid coupling in the suspended subchassis, which connects the platter through the main
bearing to the cartridge through the tonearm, is an ef fective
defense against airborne vibration.
A closed, coupled system will minimize interference from airborne energy. Rigid coupling only works at some frequencies.
It’s easy to couple low frequency bearing noise to the tonearm
but it’s far more difficult to couple high frequency vibration from
the tonearm to the platter. Even rigidly coupled, suspended
turntables must absorb and dissipate some energy at some frequencies.
Mass isolation turntables rely on high mass as a defense
against external energy and acoustic feedback. Manufacturers
who use mass isolation simply make everything bigger with the
idea that the bigger and heavier it is, the harder it will be to
Mass isolation is only partially effective. Bigger, heavier objects
are more difficult to set in motion but objects with more mass
tend to store energy at lower frequencies than lighter objects. A
heavy platter is unlikely to be moved by acoustic energy or
Turntable Physics
footfalls but a heavy chassis may store energy at low frequencies and dump that energy into the tonearm which must have
relatively low mass. Rigid coupling is more difficult to achieve
with high-mass components and any vibration in the main bearing is likely to create a signal in the cartridge. Acoustic energy
will easily affect the tonearm tube but not the rest of the
turntable system and undesirable energy that moves the cartridge body is just as bad as undesirable energy that moves the
The best turntables I’ve heard use a spring-suspended subchassis that rigidly couples the platter to the cartridge.
The Arm
The tonearm positions the cartridge above the record surface
in two planes and three directions: vertically (up-and-down),
horizontally (side-to-side) and fore-and-aft (front-to-back). The
arm must have low inertia and low resistance to up-and-down
and side-to-side movement so that the cartridge can track the
grooves on a warped record, but the arm must firmly position
the cartridge at a specific distance
from the pivot and prevent any foreand-aft or twisting movement.
Bearings are used to allow necessary movement while preventing
undesirable movement.
point that allows both vertical and horizontal movement (as well
as twisting movement that is undesirable). There are only two
parts to a unipivot bearing: a shaft with a point at the end and a
cup in which this point rests. Gravity holds the parts together
and, because the moving part of the tonearm must be relatively
low mass, coupling that relies on gravity is compromised.
Vertical arm movement when tracking a warped record will
affect cartridge azimuth in a tonearm with an offset headshell
and a unipivot bearing. (See illustration.)
Precision ball bearings provide rigid coupling while allowing
free horizontal and vertical arm movement and preventing
twisting or rocking movement which is undesirable. V ertical
bearings can be offset at the same angle as the headshell so
that azimuth remains constant throughout vertical arm movement.
Properly designed tonearms can track the record with a minimum of angular error and will have zero tracking error at two
spots (called “null points”) on the playable surface of the record
The best tonearms I’ve heard
utilize precision ball bearings
for horizontal and vertical
pivots but unipivot arms are
also common. Unipivot
bearings are much easier
and cheaper to make and
require little adjustment
for assembly but they provide inferior coupling and
allow undesirable movement. Precision ball bearings
are costly to produce and
assemble and can be easily
damaged but provide better coupling and rigidity.
A unipivot bearing has a single pivot
Turntable Physics
when geometry is adjusted correctly. Proper geometry requires
optimum offset angle and stylus overhang.
The effective length of the tonearm is the distance between the
horizontal pivot and the stylus. Overhang is the dif ference
between the effective length of the tonearm and the pivot-tospindle distance. Offset angle is the angle between the cartridge body and the tonearm tube.
For an arm of a given length, there is an optimum offset angle
and an optimum overhang distance to produce minimum tracking distortion. These figures were determined using mathematical analyses by H.G. Baerwald and others. These settings can
be adjusted only with precision instruments.
Bias compensation, which is commonly called antiskating force
or antiskate, is necessary because of cartridge offset. As the
stylus tracks the grooves, friction pulls the cartridge in the
direction of record rotation. Because the cartridge is mounted
at an angle to the tonearm tube the resulting force pulls the
tonearm inward towards the center of the record. Bias compensation is the application of a compensating force which pushes
the arm outward away from the center of the record.
Proper bias compensation should result in equal stylus pressure on the inner and outer groove walls which will center the
moving elements within the cartridge. This will minimize mistracking caused by unequal pressure and distortion caused by
the asymmetric positioning of the generator within the cartridge. Because friction varies with groove modulation, bias
compensation is always a compromise.
Vertical tracking angle (VTA) is the angle formed between the
cantilever and the record surface. Stylus rake angle (SRA) is
the difference between vertical (90 degrees from the record
surface) and a line extending from the stylus contact lines.
SRA is what we are actually adjusting when we listen for focus.
VTA changes a little when we adjust SRA but VTA is determined largely by cartridge design.
All tonearms should allow adjustments for stylus pressure, bias
compensation and VTA/SRA. Adjustment may be provided for
azimuth, which shouldn’t require adjustment if the arm is made
Rigidity is often sacrificed for adjustability. Tonearms that allow
easy adjustment of SRA, for example, can’t provide the rigidity
and coupling of arms with locking height adjustments, using
three contact points, that require tools for SRA adjustment.
Tangential tracking tonearms are supposed to eliminate tracking error. They are supposed to provide better sound by more
accurately mimicking the movement of the cutter head.
Tangential tracking tonearms create many problems in an
attempt to solve the tracking error “problem” which is a red herring in my opinion. (A red herring is a distraction from the matter at hand, which in this case is a “problem” that really doesn’t
require a solution.)
Well-designed pivoting arms actually have very small angular
errors over the normal playing surface of the record and zero
tracking error at two places (null points) within this range.
Pivoting arms can have substantially less angular error in the
vertical plane than tangential arms which are usually shorter,
and pivoting arms can be much more rigid and can provide
much better coupling than tangential arms, which must move
sideways on a carrier assembly.
There are two types of tangential tracking arms: active arms,
which are motor driven, and passive arms, which are propelled
by the stylus. Both types have a carrier mechanism that transports the arm tube and vertical pivot across the playable surface of the record.
Active arms require some kind of servo mechanism to activate
the motor that drives the arm carrier. They must move out-oftangent with the groove in order to trigger this servo. The result
is a sort of dog-tracking movement that may actually create
greater angular errors than what would occur with a welldesigned pivoting arm.
Passive tangential tracking arms must have very low resistance
to lateral movement in order to prevent stylus deflection. They
typically use air bearings or roller assemblies to allow the carrier assembly to be propelled by the stylus. Both provide poor
The best sounding tonearms I’ve heard are rigid, pivoting arms
with precision ball bearings and non-removable headshells.
Tangential tracking tonearms offer inferior performance with
added complexity, in my opinion, and they make playing
records a chore.
Turntable Physics
The Cartridge
A phono cartridge has a stylus to trace the grooves; a cantilever to conduct the stylus movement to the generator; a generator to produce electrical output; a suspension system, which
The cantilever connects the stylus, which is mounted at one
end, to the generator, which is usually mounted close to the
other end. Moving coil cartridges have generating coils, which
are positioned in a magnetic field, attached to the cantilever. In
a moving magnet or moving iron cartridge, the cantilever is
connected to a magnet or iron armature which is positioned
near fixed coils. Generator movements will closely duplicate
stylus movements.
The stylus and generator are held in position by an elastomer
suspension that acts as a fulcrum for cantilever movement and
a damper for resonance control (in most designs). This suspension component establishes the compliance of the cartridge.
Cartridge Body
The cartridge body contains all the fixed and moving parts and
precisely positions them in relation to each other. It has slots or
holes that allow the assembly to be securely mounted to the
tonearm with nuts and bolts. Cartridge bodies made from resonant materials, like wood, can add euphonic colorations to the
sound which some people enjoy.
acts as both a fulcrum and a damper, to allow the stylus and
generator to move; and a body to hold the moving and fixed
parts in precise alignment and allow the assembly to be firmly
attached to the tonearm.
The cross-section shape of playback styli has evolved from
conical to elliptical to modified elliptical—called line-contact or
some trademark name that means line-contact.
The stylus makes contact with the groove walls only at the
sides. The two small areas where the stylus touches the
groove walls are called contact patches. As stylus shapes have
evolved the contact patches have become longer top-to-bottom
and narrower front-to-back. Reducing the front-to-back dimension of the stylus contact patches raises the bandwidth of the
system by allowing the stylus to accurately track smaller
groove modulations that are closer together—representing
higher frequencies.
The sonic characteristics of phono cartridges can be modified
by electrically loading them with appropriate impedance.
Moving magnet/moving iron cartridges can be tuned with
capacitive loading, and moving coil cartridges can be tuned
with resistive loading.
In my experience, most moving coil cartridges perform best
when loaded with an impedance between 100 Ω and 500Ω. The
lower end of this range seems to provide the most electrical
damping and the higher end of the range provides the most
dynamic sound. Some cartidges are designed to work into
higher impedances up to 47k Ω.
What do you call a guy who lies on the front porch? Mat. What
do you call a 12-inch diameter flat thing with a hole in the middle, made from common materials costing mere pennies, that
sells for ridiculous sums of money? Turntable mat. Do mats
make a difference? Yes. Is this difference always an improvement in sound quality or accuracy? No.
Your ears must guide you through the turntable mat hype but
Turntable Physics
remember these facts: unless the mat is thinner in the label
area where records are thicker, it won’t even touch the playable
surface of the record. A mat that is the same thickness all the
way across will support the record only in the label area and at
the outer edge—the playable surface will be hanging in thin air.
A soft rubber mat without a record clamp will decouple the
record from the platter. If you have a ringy glass platter this
may be desirable. If you have a well-coupled suspended
turntable it probably will be undesirable.
Resonant Systems
A number of resonant systems have been described in this article.
frequencies must
be carefully positioned for
include the
suspension resonance, the tonearm/cartridge resonance, and the cantilever
Suspension Resonance
Spring suspension systems should bounce (resonate) at 3-4Hz.
This is about as low as you can go while retaining practical stability. A suspension with a resonance of 3-4Hz will block most
frequencies above 4Hz providing good acoustic isolation, but
won’t bounce too far or too long as it might with a resonant frequency lower than 3Hz. The resonant frequency of the suspension is established by the mass of the suspended parts and the
stiffness of the springs and is not user-adjustable in most
Cartridge/Tonearm Resonance
The moving parts of a phono cartridge include the stylus, cantilever and the generator element, which may be a tiny piece of
magnet or iron or a set of moving coils. The moving parts are
suspended by an elastic fulcrum/damper. The compliance of
the cartridge suspension and the effective combined mass of
the cartridge and tonearm will establish one of the most important resonant frequencies in the turntable system. This frequency should be as low as possible while remaining above the resonant frequency of the suspension and most warps.
The cartridge/tonearm resonance is very important because it
sets the low frequency limit of the turntable system and determines the character of the bass to some extent. This frequency
is established by selecting a proper match of cartridge compliance and tonearm mass and can be fine-tuned by varying
mass slightly.
As a general rule, a low compliance cartridge should be mounted in a high mass tonearm and a high compliance cartridge
needs a low mass tonearm. Playing a test record with a 5-20Hz
sweep tone and graphing the results is the best way to determine the actual resonant frequency. Very small changes in
mass, like the difference between aluminum and stainless steel
cartridge mounting hardware, can alter resonant frequency.
Expert opinions differ on the optimum frequency for
cartridge/tonearm resonance. The range usually recommended
is 6-15Hz but I will state without equivocation that the magic
number is 8Hz. Go higher than that and you’ll get fatter bass
with less definition. Go much lower and the stylus will jump out
of the groove if it encounters the slightest warp.
Cantilever Resonance
The flexibility of the cantilever tube (or rod) and the mass of
the stylus and the generator will establish the high frequency
resonance of the cartridge. At this frequency there will be a
peak in output and above this frequency the output of the cartridge will begin to fall. This frequency should be as high as
possible and you can exercise some control over it with your
In the old days cantilevers were made from aluminum tubes
and styli were made from big chunks of diamond and mounted
in little cups at the end of the cantilever. Lower-priced cartridges are still made this way. Today’s better stylus tips are
ground on tiny diamonds and “nude” mounted on cantilevers
made from extremely stiff materials like boron. This reduces
moving mass and creates a stiffer spring, which equates to a
higher resonant frequency and broader bandwidth. Sometimes
technology isn’t just hype. Sometimes you can really buy better
performance by spending more. APJ
Brooks Berdan
Brooks Berdan is well
known in the high-end
audio industry. He
began applying his
mechanical engineering
skills to audio components in the 1970s and
his meticulous turntable
set-ups have established a benchmark to
which others can only
aspire. Over the years
Brooks has created a number of innovative modifications to
existing turntable and tonearm designs and inspired new
design concepts that have helped to raise the standards of
analog performance. Today he is one of our most experienced
vinyl advocates. Manufacturers consult with him on new
designs and audiophiles from all over the world seek his services.
Brooks Berdan and I used to spend most Saturday mornings
riding nearly-legal motorcycles at ludicrous speed through the
mountains and canyons above his retail store, Brooks Berdan,
Ltd., in Monrovia, California. We’d put 350 miles on the odometers before noon and then go to work. I’ve known him for a long
time and have come to respect his perfectionist nature. This is
a man who simply won’t settle for “good enough.”
Brooks has hands-on experience with virtually every brand and
type of turntable, arm and cartridge because his services are
available to everyone, not just those who purchase the products he recommends and sells. I asked if he would provide a
rudimentary explanation of turntable set-up procedures for
Journal readers and he graciously agreed. This information will
be helpful if you want to try to do the job yourself and it will
increase your understanding of turntable systems even if you
have no desire to actually work on them.
If you choose to hire an expert to set up your turntable, a little
knowledge about the important adjustments and how they’re
done will help you pick the right technician to do the job.
Knowing what’s involved will help assure that you get what
you’ve paid for.
Brooks Berdan and I met at the Audio Perfectionist laboratories (my house) in February 2003 to discuss the subject of high
performance turntable systems for this article. Here’s a transcript of our conversation.
Brooks, do you have any caveats for readers who want to set
up their own turntable?
Yes, there are some points that I would like to stress before we
begin. First, it is essential that a turntable be absolutely level
during set up and operation. If the turntable is even slightly tilted, in any direction, all adjustments will be compromised and
performance will be negatively impacted. Second, cartridges
require 12-15 hours of mechanical break-in to allow the suspension to settle. (Sound will continue to change over a longer
period.) A good dealer will break-in a phono cartridge before
final adjustments are performed. Readers who set up their own
turntables should readjust stylus pressure and VT A/SRA and
recheck tonearm geometry after 12-15 hours of play. Third,
some critical adjustments are best done by ear and when I talk
about adjusting for focus, et cetera, I’m assuming the reader
has a properly functioning audio system that allows dif ferences
to be heard. Don’t try to adjust the turntable to compensate for
other flaws in system performance. Fix the flaws instead.
Remember, all turntable set-up errors are cumulative because
each adjustment affects all other adjustments. Make each
adjustment in the proper sequence and do each one correctly!
That’s good advice. How do you prepare?
A firm, level work surface is necessary. Basic hand tools are
required along with any special tools needed for tonearm or
suspension adjustments. Special tools are usually furnished
with the turntable or tonearm. A precision alignment gauge
must be used to align the cartridge and set stylus overhang. A
carpenter’s level is essential. A straight edge level works much
better than a round bubble level.
When the work area is ready and the tools are collected how
do you begin the set-up?
I start assembling a mass-isolated table by leveling the platter.
The feet should be adjusted to be as close to the plinth as possible and locking nuts tightened to eliminate vibration and
movement. Next verify that the plinth is level.
Brooks Berdan
If the platter is level and the plinth or arm mounting surface is
not level, you’ve got a serious problem. Fix it before you proceed. All level checks should be done in two perpendicular
directions—side-to-side and front-to-back.
That covers the first stage of set-up for tables without spring
Are suspended turntables assembled differently?
Many suspended
tables, like
must be elevated for
access to
the chassis adjustments. These tables should be mounted in a
secure set-up jig that allows you to work above and below the
plinth. The set-up jig must be level.
Again the first step is to level the platter and verify that the arm
mounting surface is parallel to the platter surface. If the arm
mounting surface is not absolutely parallel to the platter surface
the situation must be corrected. The tonearm will not work correctly if there is even a small deviation from parallel. Fix the
problem or replace the defective part. Do not use shims in an
attempt to compensate for non-parallel surfaces!
After we’re sure that everything is level and true, what’s next?
At this point I install the cartridge in the tonearm with the bolts
positioned at the midpoint of the headshell slots, mount the
tonearm on the turntable and pre-adjust the counterweight.
Before suspension adjustments begin, the tonearm cable and
belt should be installed and the tonearm should be placed in
the midpoint of travel over the playable surface of the record.
Then the suspension can be leveled.
Next orient the springs for pistonic bounce—the subchassis
should bounce straight up and down with no wobble or sideways oscillation—and test using gentle pressure to bounce the
suspension. Make sure that everything associated with the
springs is straight and true. If there are bolts in the springs
make sure they are straight. Remove any burrs from rubber
grommets and seat the grommets in chassis holes. Some
grommets, like those used with the Linn Sondek, are eccentric
to allow the position of the subchassis to be adjusted. The
position should be adjusted so that there is clearance between
the arm board and plinth all around.
It’s important to understand that the suspension is critical to
isolation. It must bounce freely at 3-4 Hz or lower. It must
bounce straight up and down. Except for the tonearm cable
and belt, there must be no mechanical contact that might allow
vibration to bypass the springs. You must use the correct belt
and tonearm cable. Incorrect belt tension can pull the subchassis to one side and defeat the suspension. A tonearm cable
that is too stiff can alter the bounce frequency and may bypass
the suspension.
After the turntable chassis is adjusted we can begin with the
critical adjustment of the tonearm and cartridge, right?
That’s correct. There are five basic adjustments and they are
interactive so I do them all twice. The first time through I get
things in the ballpark to minimize cumulative errors and then I
go back and set everything perfectly. Let me give you some
examples to show why this is very important.
You must get the VTA/SRA (vertical tracking angle/stylus rake
angle; see Turntable Physics in this issue) close before you
adjust overhang. If the back of the tonearm is very high, for
example, it will affect the overhang dimension. Overhang must
be adjusted before stylus pressure because moving the cartridge back and forth to adjust overhang will increase or
decrease stylus pressure. Changing stylus pressure will force
the cantilever upward or downward affecting VTA/SRA, etc.
Give us the order in which these adjustments should be performed.
I start by adjusting the height of the back of the tonearm until
the top of the cartridge body is parallel with the record surface
or the cartridge is slightly low at the rear. This gets the
VTA/SRA near the correct setting. I do this by sighting on the
top of the cartridge body, not the top of the tonearm tube which
may or may not be parallel to the top of the cartridge body. At
this point the counterweight should be installed and positioned
to provide some stylus pressure. I start on the light side of the
range of acceptable stylus pressure and proceed with the
Brooks Berdan
adjustments that follow. Final adjustment of stylus pressure will
occur later.
the magnet or iron element will be centered in relation to the
coils for the most linear operation.
After these preliminary adjustments are completed I adjust
azimuth. This can be accomplished by observing waveforms on
an oscilloscope or by dropping the stylus on a mirror which
makes any angular error easier to see by visually doubling the
error. If the tonearm does not allow azimuth adjustment do not
try to compensate for error by shimming the cartridge or the
arm. Accept things as they are or replace the arm or cartridge.
Antiskate (bias) force can be adjusted using a test record and
an oscilloscope. This takes some experience. If bias is set too
high based on mistracking of highly modulated grooves it may
cause mistracking on grooves with lower modulation levels like
those found on most recordings of music. Antiskate can also be
adjusted by playing records with high levels of modulation and
listening for mistracking. If you hear mistracking primarily in the
right channel, increase antiskate force. If you hear mistracking
primarily in the left channel, decrease antiskate force.
After initial settings have been accomplished I adjust overhang
and offset angle. This adjustment must be performed with the
aid of a precision mechanical instrument. If you don’t have a
gauge, find someone who does. I use the null points developed
by Baerwald, which have become industry standards.
Now stylus
force) can be
set and antiskate (bias)
force applied.
If the cartridge manufacturer has a
stylus pressure, like 1.7 grams, start there. If the recommended setting
offers a range of pressure, like 1.75-2.25 grams, start at the
high end of the range. An oscilloscope can be used to observe
mistracking on a highly modulated test record but ultimately
you must install the turntable in an audio system and listen
while making fine adjustments. Gradually reduce stylus pressure and listen for image focus using a simple recording of a
solo instrument or voice. Set the final pressure in the middle of
the range where the best image focus and most realistic
soundstage size is achieved.
Stylus pressure is a critical adjustment because it does more
than simply press the stylus to the groove walls. The correct
stylus pressure will center the moving element within the cartridge in relation to the fixed elements. In the case of a moving
coil cartridge, that means centering the coils in the pole piece.
With moving magnet or moving iron cartridges it assures that
Antiskate or bias force cannot be set using a grooveless record
because the skating force you are compensating for is created
mainly by friction in the groove and there is far less friction
when the stylus is gliding across a smooth record surface with
no grooves.
Final adjustment of VTA/SRA comes last—not because this
adjustment is unimportant but because all other adjustments
must be perfect before VTA/SRA can be set correctly. This
adjustment is commonly referred to as setting the vertical tracking angle, but what we are actually optimizing is stylus rake
angle. (See Turntable Physics in this issue.)
First listen for image focus and determine the minimum and
maximum settings. Fine-tune within this range of adjustment.
Next listen for harmonics and adjust between dull and bright
sound. This adjustment is always a compromise and the setting
will change when you play records of different thicknesses or
records that were cut at different angles.
After these adjustments have been performed, are we done?
After I have done all the adjustments described above, I go back
and do them all again, in sequence, to correct for minor errors
that have occurred due to interaction between adjustments, and I
tighten everything down so the settings will stay set. VTA/SRA
and stylus pressure adjustments are interactive and may require
some tinkering to arrive at the best compromise setting.
Are there any special considerations for tangential tracking
Tangential tracking arms must be adjusted so that the arm car-
Blue Collar Confessions
rier is absolutely level across the entire range of travel. The
cartridge should be positioned for zero overhang and the arm
adjusted to be parallel to the record surface and tangential to
the record grooves. The cartridge should remain perpendicular
to the record radius across the entire range of travel. Stylus
pressure and VTA/SRA are set in the same manner as for pivoting arms.
Some tonearms offer a choice of counterweights. With a pivoting arm a heavier counterweight positioned closer to the pivot
is preferable to minimize inertia. With tangential tracking arms
a lighter weight positioned farther from the pivot is usually better to facilitate lateral arm movement.
Tangential tracking arms, particularly the passive type, can be
affected by lead dressing. Cue the arm up and down in the
midpoint of travel and observe whether the arm tends to move
inward or outward and dress the cartridge leads for minimum
effect. APJ
High Cost of Hi-Fi
It’s easy for those of us with middle class incomes to become
intimidated by the high prices of high-end audio components.
The Journal has tried to dispel the myth that true high-end performance cannot be achieved without spending a king’s ransom
on an audio system. While good—even great—performing components are available for less than that, outstanding equipment
remains relatively expensive for those of us who work for wages.
Dealers are reluctant to discuss alternatives to purchasing new
components at retail because they mistakenly believe that this
may cost them business. In fact, a healthy used market stimulates business by providing a means for new customers to
embrace our hobby. These newbies can get started with used or
demo components they otherwise could not afford and they are
likely to purchase top-of-the-line upgrade components one at a
time when they become aware of the deficiencies of the bargains
that got them started. A healthy used market also allows the
more affluent buyer to recoup some of the investment in a component that, for whatever reason, has proven to be personally
unsatisfying. This encourages that affluent customer to reinvest,
perpetuating the business.
I met Chris Fitzgerald when he attended an AP Seminar. Chris
has managed to assemble an outstanding system for a remarkably small investment. I asked him to share his experience with
other Journal readers who may be working with budget constraints.Here’s his story. —RLH
I really love music but I’m a
family man with financial
obligations that preclude
huge expenditures on highend audio components.
Even so, I have managed,
without undue financial
sacrifice, to assemble a
remarkably satisfying home
entertainment system by
purchasing used and demo
equipment. The total cost of my high-end system was less than
half the combined retail price of the individual components.
I am a long-time Journal subscriber and an equipment junkie. I
attended Audio Perfectionist seminars in August and
November (2002). That opportunity changed my perspective on
home music reproduction, clarifying much of my earlier confusion about how to attain an accurate and honest sounding setup in my own home.
The AP seminars, of course, included discussions on equipment selection in various price ranges. With this information,
and some creative financing, I’ve managed to put together a
system that is top-notch. I assembled a system with a retail
value of somewhere around $35,000 for a total expenditure of
only about $14,000.
Prior to the AP seminar, I already owned what I thought was
first-rate equipment—made by TAG McLaren, B&W, Theta,
Definitive Tech, Sony, and Mitsubishi—along with an assortment of cables. I had a system that I thought was very dynamic
and for me very exciting. As I said, I thought I was doing well
until my first visit to Richard’s house. After the seminar, armed
with my newfound knowledge and a desire to achieve a somewhat identical sound to Richard’s, I did what I told myself I
wouldn’t do for a long time. I began replacing equipment. Much
to the dismay of my wonderful wife, I sold everything piece by
piece, again.
After hearing Richard’s system, I had a much better idea of
what type of equipment to obtain. I wanted to have my system
built on the same principles that Richard recommends. A set-up
Blue Collar Confessions
“Frequently, the sellers are
people who tend to require the
latest, prettiest, or most prestigious toys; I have bought much
of my good equipment from
such people.”
consisting of time- and phase-accurate speakers, fully balanced amplification with zero feedback, and—something completely new to me—a solid turntable.
Richard has asked me to share with other Journal readers my
methods for attaining good equipment at reasonable prices.
First of all, buying used and demo equipment has allowed me
to own better components than I’d be able to afford if purchased new. To achieve this, I buy from internet sites, local
dealers, and other contacts in the retail business. I have
learned through Richard’s seminars how to evaluate equipment
based on my needs: what features to look for, quality of workmanship, manufacturer’s reputation, etc. I use the internet
extensively to research the equipment I am interested in. I get
involved in chat rooms on various AV sites to hear dif ferent
points of view on equipment and, lately, I have read a lot about
The internet has all of the tools that a thrifty AV shopper needs.
Frequently, the sellers are people who tend to require the latest, prettiest, or most prestigious toys; I have bought much of
my good equipment from such people. To give you an example,
I bought my Theta Dreadnaught amplifier from a lawyer on the
east coast. He had the piece listed on an internet site for several months and, by the time I contacted him, he was fairly
frustrated that it hadn’t generated more interest. He was dissatisfied with it, he said, and just wanted it out of his living room. It
turned out that it had a small ding in the top corner of the faceplate. Since I’m generally more interested in how a component
performs than what it looks like, I was able to get this fine amp
at a great discounted price.
I recently bought a Linn LP-12 from a seller in Florida who had
given up on LPs. The ad included a very basic description and
some poor quality pictures. This turntable had a broken cartridge and a broken lock on the tonearm rest. Knowing I could
have these parts fixed for very little money, I took a closer look.
It had an Ekos tonearm, which sells for $1,000 used. I got the
whole package for $1,000! When I questioned the owner about
the features on the table he knew nothing about it. He admitted
to me that he just went to his local dealer and bought it without
even a demo.
My system now includes Vandersteen 2CE Signature speakers,
a VCC1 center channel speaker, VSM1 surround speakers, a
2WQ subwoofer, a Theta Dreadnaught amplifier, a Classe SSP75 surround processor, an Ayre
K3x preamp, a
Linn LP12
with Ekos
and Troika cartridge,
player, and Audioquest Volcano, Anaconda, and Python
cables—all of which were purchased used on the internet,
except the Classe SSP-75, Sony DVP, Vandersteen surrounds
and Audioquest cables, which were purchased from dealers. All
of the components were demos or closeouts, or purchased
used on the internet. Several pieces have some sort of minor
cosmetic defect—i.e. a scratch on the cabinetry—which doesn’t
affect performance.
If you’re willing to look beyond the superficial, you, too, can
obtain first-rate equipment at killer prices. Stay open to buying
used and demo pieces from local dealers. Do your homework,
research the components and, most of all, know what you
want. Be ready to buy the components when they come up—
timing is everything!
In a short period of time I have been able to go from a good
system to an outstanding system that is everything I ever imagined it would be, one that is time- and phase-accurate, and
fully balanced. I just had to overlook two small details: none of
it matches, and every piece has scratches. Until next time,
happy hunting. APJ
Joe Harley
these things to Direct Stream Digital recordings on hard disc
and to the SACD discs made from these DSD hard disc recordings.
t the 2003 Consumer Electronics Show (CES), as in years
past, I found many high-end audio exhibitors demonstrating the capabilities of their products with analog, digital and
SACD recordings produced by Joe Harley. Harley and his engineer/partner Michael C. Ross have achieved respect in the
finicky high-end audio industry for their abilities to recognize
good music and to capture the essence of that music in their
recordings so that it can be reproduced in our homes (and at
hi-fi shows) with great sound and the emotional message
Joe Harley and I have been friends for many years. I thought
that he would make a good candidate for an interview in the
Audio Perfectionist Journal for a number of reasons.
Joe makes great sounding recordings of music that you want to
listen to, not use just to show off your system. He is a music
lover with audiophile sensibilities who is experienced in every
area of high-end audio. His knowledge of high-end audio reproduction assures that his music productions are held to high
standards—Joe knows the capability of high resolution play-
Joe, your recordings have become a staple source for music
lovers who care more about content than sound quality, and for
high-end audio enthusiasts who demand excellent sound.
You’ve managed to capture the musical message and make
recordings that sound good.
Tell us about yourself and the history of Harley Music
Productions. How did you start? What motivates you?
I had been working with Bill Low at AudioQuest since 1983,
helping to establish the company and set up our dealer/distributor network. In 1990 Bill and I had the idea of doing a “one
off” promotional recording using AQ cables. This record turned
out to be the Robert Lucas classic Using Man Blues. The
record was a sensation in blues circles and helped establish
Lucas as a major figure in the blues world. It also generated a
lot of attention in the audiophile community with Corey
Greenberg, then of Stereophile, naming it the “audiophile
recording of the century.” Thinking back on the recording now, I
really had no idea at the time what “production” meant. But I
loved the music, appreciated good sound and basically just
flew by the seat of my pants.
We did the early AQ recordings with Kavi Alexander and his
fabulous Blumlein/DeParavacini set-up. [Blumlein conceived
the crossed/coincident microphone technique and T im
DeParavacini was responsible for the modifications to the
recording electronics and tape deck.] I still think it is an excellent way to record, particularly solo and small acoustic ensembles.
Joe & Michael
back systems and strives to capture the very best sound on his
recordings. He has compared the sound of live music to the
sound of a monitoring system playing a direct mic feed, giving
him an objective basis for comparisons of recording technologies. He has compared analog tape recordings to a direct mic
feed and to the sound of live musicians in the studio. He has
compared analog master tape recordings to the compact discs
and vinyl records made from those tapes. He has compared all
I have enjoyed learning about Joe’s experiences and I’m sure
you will, too. Here are my questions and Joe Harley’s responses:
I became totally fascinated with trying to capture the best performances…the musical juice…on tape. I loved the simple mic
technique for certain projects, but I began to realize that there
were obvious limitations to the approach, particularly when the
group got larger or drums or electric instruments were involved.
I would end up telling everyone to play more quietly, or move
an inch here and there. This kind of thing can really put a
Joe Harley
damper on the creative energy in a session. Musicians end up
having to worry about more than their own artistic expression. I
decided that for much of the music I wanted to record, a dif ferent approach would be needed. I focused on trying to maximize
the traditional studio experience. I insist that the sonics serve
the music, not the other way around.
Meanwhile, AudioQuest Music was born as a distinct entity
separate from the cable company. I wanted a label to make
records that put musical merit as a first priority. I was extremely
fortunate to find a few engineers, and in particular Michael C.
Ross, who could help bring this vision to fruition. At the time,
some so-called audiophile labels had a reputation for making
great sounding but musically tepid recordings. I wanted to
make recordings that could stand on their own in the blues and
jazz world.
AQM jazz recordings featured such artists as Grover
Washington Jr., John Abercrombie, Bennie Wallace, Kenny
Barron and James Newton. The AQM blues roster found success with such artists as Mighty Sam McClain, Ronnie Earl,
Terry Evans, Ry Cooder and Doug MacLeod.
In 1996 I began to
do production work
for other labels,
GrooveNote, Enja,
Telarc and ECM in
addition to
AudioQuest Music.
I left AudioQuest
and formed Harley
Music Productions
at this time, working with such artists as Jacintha, Charles Lloyd, Brad Mehldau,
Eden Atwood, Anthony Wilson and Robert Lockwood Jr. The
AQM label was eventually sold to Valley Entertainment in 2000.
I also began a long association with the JVC/XRCD program at
this time, serving as Creative Director.
How do you choose projects?
Projects come to me in a variety of ways. Sometimes a label
will assign a project to me but more often I will bring an artist
whose work I particularly admire to the attention of a label. I
have also had a number of artist referrals, meaning that someone I’ve worked with will recommend me to another artist.
The artist and I will come up with a project proposal and I’ll
draft a budget. If the label approves, we start the wheels turning for a session.
What projects are you proudest of?
Difficult question. Anytime the music is completely happening
and you’re getting it all on tape or hard drive properly…those
are the magic times, when you know that all the effort is worth it.
Having said that, there are a few projects that stand out for me.
I’ve worked on the last three Charles Lloyd albums for ECM.
The first two included a dream band for a jazz lover like
myself….Brad Mehldau on piano, John Abercrombie on guitar,
Larry Grenedier on bass, Charles of course on tenor sax and
the late, great Billy Higgins on drums. It was one of Billy’s last
sessions. His health was failing, he was so weak. But when he
would get on the drums he would totally transform and be Billy
Higgins at his finest! It was something to witness. He would
look at his body and say, “This is all gone, it’s all spirit now .”
What an amazing man.
I’ve worked with Ry Cooder a number of times on Terry Evans’
projects. He really understands that it’s the soul of the music
that drives the thing, not trying to make it perfect. Ry always
wants to get the thing on tape when the band is right on the
cusp of learning the material, when the element of surprise and
wonder is fresh. He is a master of texture and sound.
I have a special
fondness for several artists who
are lesser known
for whatever reason, but who
should be on any
best of the best
list. Tenor sax
player Bennie
Wallace, vocalist
Eden Atwood, guitarist Anthony Wilson… these artists deserve to have much
wider recognition.
Joe Harley
There will always be a special spot for R & B master Mighty
Sam McClain. Just a few months before we recorded Sam’s
first comeback album in 1992, he was living on the streets of
Boston and eating out of garbage cans. But his talent was
intact. The record came out and was a sensation. Sam and I
went on to make 7 albums together. Last year one of the tracks
from his most recent album was used as one of the recurring
themes on the Ally McBeal show. Sam bought a large tour bus
on the proceeds from that.
What equipment do you use to make original recordings?
I like to use vintage consoles and vintage mics….they just have
a rightness to me that is missing in so much modern gear. For
me that means Neve consoles like the 8038 and 8078. At one
of my favorite studios, OceanWay Studio B in Hollywood, there
is a custom console that I love. It’s mostly passive and nearly
transparent. For microphones, I like the vintage Neumann’ s…
the 269, M-49, M-50, U-47s. AKG C-12s and C-12As are used
frequently on my productions, most often for piano and drum
overheads. The Sony C-55P is an amazing mic, especially for
string bass and guitar. You pretty much have to record at one
of Allen Sides’ facilities (OceanWay and Record One) to use
them, though, since he has corned the market on these babies.
When you record in the studio are you able to monitor with
components that provide a reasonable facsimile of the live
sound? Does the microphone feed played through the speakers
in the studio sound like the live performance?
This is a very good question. The answer used to be “not
always.” In fact when I first began working at OceanWay I was
panic stricken since their own highly touted monitors sounded
bright and boomy to me. Dynamic as hell but not a tonal balance I could relate to. I would drive back to San Clemente to
check balances on my own system (various combinations of
Vandersteens and ARC components). Then my engineer, Mike
Ross, brought in some Tannoy monitors that had been modified
by the Mastering Lab. Bingo! Here were monitors that were
robust enough to live in a studio but had a tonal balance that I
felt totally at home with. For many years now we’ve been dragging them to every session….and now we have 5 to cover us
for surround sessions.
When you compare a direct mic feed to the analog tape
recording of that feed what differences do you hear? How
much is lost in the basic recording process?
Again, the live mic feed
is more direct and “open”
than even the best analog tape. It isn’t a major
difference but it is very
easy to pick up on. The
sound coming back on
the analog tape is rounder and slightly blunted by
For years I recorded solely to analog… no digital system I
heard could come close to competing with the natural warmth
and bloom of analog tape. My machines of choice were made
by Ampex… the ATR1002 and 1024 machines.
Then, in early 1999 I had my first exposure to Sony’s Direct
Stream Digital (DSD) format. It changed everything I had previously thought about digital. It rocked my world. I remember
very clearly looking over at Mike Ross as we made our first
comparison of the direct mic feed to the DSD hard drive to our
trusty tweaked ATR 1002 analog machine. The DSD sounded
like the direct mic feed! For the first time I had a clear handle
on what analog tape DIDN’T do right. The tape still sounded
great, beautiful as always. But now I could clearly hear the
“rounding” artifacts and loss of transient impact. The DSD
sounded so much like the direct feed that I would get confused
going back and forth between them….they both sounded so
Since then I have recorded or mixed almost exclusively on
DSD and have done a number of projects in 5.1 DSD.
When you compare the original master tape recording to the
commercial vinyl records and CDs made from those recordings,
what musical information seems to be missing? Which provides
a better representation of the master tape, vinyl or compact
disc? Which one best conveys the musical message to you?
I feel that you have to look at LP and CD separately. There
really is no comparison. Yes, it is possible that a bad LP pressing will be beat by a superb mastering of a CD. It’s possible.
But for the most part LP retains much more of the critical information that makes you feel the music on an emotional level.
That thing that makes you tap your foot, nod your head, cry,
Joe Harley
laugh…whatever it is the artist is intending for you to feel...is
just so much more alive on vinyl. CD comes off as kind of dead
by comparison. You turn the volume up trying to make something happen, but the deadness just gets louder. You find your
mind wandering, you start reading the paper, doing other
things. After a while you turn it off. Your brain is being asked to
fill in the blanks that 16-bit/44.1kHz can’t provide. So listening
to music becomes work! Music is supposed to be relief from
But what you can do is to tastefully go for a more immersive
experience. You can do this by expanding the front stage left to
right, which now also opens up the layering within that stage.
Think of it as a horseshoe with the listener’s position at the
center opening of the horseshoe. Then you bring a sense of
ambience, the room mics, into the rears. It can be very ef fective and involving. The stereo mix just is not as much fun to listen to!
What projects have you completed recently?
Tell us about your experiences with Direct Stream Digital
recording and SACD discs. How does a DSD master recording
on hard disc compare to the mic feed?
DSD is a major advance in my opinion. It is the closest archival
medium we have for capturing the sound of the direct mic feed.
In fact, if we are talking about the sound of the DSD hard disc,
the sound is so close as to be nearly indistinguishable. All the
life, the musical “juice” is retained on DSD.
I just finished another surround DSD mix for GrooveNote with
the popular Asian singer Jacintha. I have also been supervising
DSD transfers of older Concord material, again for SACD
release on the GrooveNote label. A great Ray Brown album—
Soular Energy—and LA Four album have already been
released. Next week we go in to do the DSD transfers for Stan
Getz and Rosemary Clooney albums.
What other projects are coming up?
How does a commercial SACD disc compare to the original
DSD recording on hard disc?
You ask a very interesting question about how the information
is retained once the data is processed for commercial SACD
disc. The answer is it mostly is… and more of it is retained now
than even a year ago. Those of us involved with DSD recording
began to notice that when we backed up the hard drive onto an
AIT tape something happened. Something was lost. Sony
began to investigate this….minds like Ed Meitner got involved.
Now what you get back on the commercial disc is much more
like the original hard drive. It was good to begin with and it’s
getting better all the time.
I know that you have done some multichannel projects. Do you
think that multichannel offers an improvement in musical fidelity?
Yes, at its best I believe it can. Of course there’s also the
opportunity for rampant abuse as we’ve all heard. I had an
absurd mix just the other day by a guy who should know better.
He had half the drum kit front right, and the high hat and snare
rear left. Whatever could he have been thinking? This totally
discombobulates the feel of the rhythm section for the sake of
cheap carnival tricks.
Next month I will start my first project for DVD-A release for
DTS. This will be a surround mix of an album originally
released on Blue Note by the band Medjeski, Martin and W ood.
I love the band’s music so this is one I’m really looking forward
to. For this one forget everything I said about surround mixing… this music has no rules; neither will my mixing!
In April we will be recording the wonderful vocalist Eden
Atwood again. She is the real thing…she makes you feel the
song. We’ll have James Moody on board and some other great
In May, I’ll be in New York to record Bennie Wallace in a trio
with the great Kenny Barron on piano and Eddie Gomez on
bass. Later this year I’ll be going back in for another Charles
Lloyd album on ECM.
What does the future hold for Harley Music Productions?
In October of last year I rejoined AudioQuest after a hiatus of
six years. It’s great to be back working with Bill Low again. As
you can see, part of our deal was that I keep my outside music
clients. When a project comes along that I want to do, I go do
it. Music and audio… my two addictions. What could be better
than that? APJ
intended to begin this section of this Journal with an article
that would help to clarify the confusing world of digital audio.
I wanted to replace the rhetoric and hyperbole with simple
explanations that readers could understand. Obtaining the necessary background information, however, proved to be more
difficult than I had anticipated. My research led me to industry
“experts” who either couldn’t or wouldn’t provide me with
specific data explaining the actual processes involved. I
encountered smoke and mirrors almost everywhere I turned for
Many of the “authorities” I consulted seem to prefer withholding
the truth from me and other consumers. Instead they offer marketing slogans and meaningless terminology along with
definitions that make no sense. Why, you might ask, do they
want consumers to be confused and bewildered about what’s
really going on in the digital audio world? Do they feel that providing candid explanations is futile because we simply wouldn’t
understand? Are they so much smarter than we are?
Is “upsampling” really different from oversampling? Is pulse
density modulation really different from delta-sigma modulation? The answer to both questions appears to be no, but manufacturers and industry pundits who have “hung their hats” on
new terminology, which doesn’t necessarily describe new technology, continue to protest when confronted by evidence and
asked for specific information about what distinguishes their
“unique” processes from common techniques that are better
known and more easily understood.
Differentiating Your Product
Those of us who are familiar with computers know that often
there is little to distinguish between the various brands. A
dozen or so manufacturers assemble machines around CPU
chips made by a small number of companies mounted on
mother boards made by a few others. Add video and sound
cards from a limited number of sources, and various other
specific components like disc drives, and you can assemble a
machine that runs the same software and works pretty much
like all the others. Digital audio components at the low end of
the market are similarly indistinguishable. At the high end, consumers want something different to justify the extra expenditure
that high-end audio products demand. Manufacturers often
offer rhetoric instead of innovation to differentiate their products
from all the rest. They particularly want to distance themselves
from competing products that perform the same function but
cost less.
Digital audio is a complex and confusing subject that won’t be
fully explained by a few thousand words even if I could fully
explain a subject that I can’t claim to fully understand. My original goal won’t be completely accomplished in this Journal but I
won’t give up. Consider this an opening salvo in a continuing
battle to reveal the truth to our readers as we discover it. I am
fairly confident that the information presented here is essentially correct but I will be happy to modify or rescind any statements that prove to be misleading or wrong. I’ll continue to
learn and present what I discover to you.
The Analog World
We live in an analog world. Daylight fades continuously into
darkness and the howl of a lone wolf startles the night and
fades continuously into silence. We could record representative
samples of the sunset as it occurs with a series of frames
(samples) on movie film. Projecting the movie at 24 frames per
second would provide an excellent visual simulation of the
actual continuous event. We could record the wolf’s howl on
analog tape or convert the analog waveform to a series of samples in the form of a timed digital code which can be stored on
a digital medium. The analog tape will provide a remarkably
accurate reproduction of the live sound, but how about the digitally coded recording? Can a series of timed samples allow us
to hear more than a pale facsimile of the original event?
Converting a continuous event into a digital code consisting of
timed samples provides many advantages for storage and
transmission but a sufficient number of samples must be taken,
and sophisticated coding techniques must be employed, in
order to preserve every nuance of the live event. The digital
coding techniques of the past have convinced many audiophiles that digital audio is inherently bad. Newer, higher resolution digital formats provide hope for an audio future that is both
convenient and musically satisfying.
Can Digital Audio Work?
(images and aliases) from being introduced into the signal.
Humans perceive the events in an analog world using cells that
perform in a digital manner—these cells seem to be either off
or on. Neurons, which are the basic functional elements of our
nervous systems, communicate with impulses that pass
through synapses connecting them. These components of the
human nervous system seem to work like the components in
digital computers.
Bandwidth limiting is accomplished by filtering and the filtering
process is accompanied by problems. Filters tend to distort fast
transients in music because filters don’t respond to impulses in
an ideal manner. Filters ring—continue to oscillate after, or in
the case of digital filters before and after, the transient.
We have an enormous number of these cells giving us the
capability to perceive events in an almost continuous manner.
Our eyes can be fooled by movies that project still images 24
times per second but our hearing is far more sensitive. While
many people find that compact discs, which sample 44,100
time per second, can’t provide a completely satisfying musical
experience there is little question that digital can work because
it seems to work in our own bodies. Perhaps what we need is a
digital audio system with higher resolution capability.
Just what is a digital signal and how does it differ from an analog signal? How is the resolution of a digital signal determined?
How do the new high res digital formats differ from the digital
coding on the familiar compact disc?
Digital Signals
A digital signal is composed of sequential samples that represent numerical quantities at an instant in time as opposed to an
analog signal, which has continuous physical variables.
The air pressure variations that we perceive as sound are continuously variable. These continuously variable analog “sound
waves” must be sampled periodically and converted into a digital code in order to be stored on a digital medium. This digital
code must be reconverted to an analog signal for playback. An
analog-to-digital converter is called an ADC and a digital-toanalog converter is called a DAC. The sampling theorem
defines the requirements for accomplishing the conversion from
The signal is low-pass filtered before sampling (analog-to-digital conversion) by an anti-aliasing filter. During reconstruction
(digital-to-analog conversion) an anti-imaging filter is applied.
The method and application of this filtering has a significant
effect on sound quality.
A digital code consisting of sequential samples that represent
the amplitude of the analog signal at specific instances in time,
or the change in amplitude between one sample and the next,
will be created during analog-to-digital conversion. The digital
code will consist of many samples representing many instances
in time. With linear pulse code modulation (LPCM), each digital
sample will have a numerical value that represents the amplitude of the analog signal at the time of the sample. In the
Direct Stream Digital (DSD) process, the density of pulses will
represent the analog signal.
A large number of digital samples will provide a good representation of the analog waveform over a period of time. Increasing
the number of samples in a given time period, or increasing the
size of each sample, can provide a better representation of the
analog waveform.
The digital code consists of a series of ones and zeros—actually a series of pulses occurring in a time frame. These pulses
can represent one of two states. The presence of a pulse can
represent one state, and the absence of a pulse can represent
the other state. These two states can represent binary numbers. What, you might ask, are binary numbers?
Binary Numbers
The sampling theorem, developed primarily by Nyquist and
Shannon at Bell Laboratories, states that a continuous analog
signal can be converted into a stream of timed samples without
loss of information if the analog signal is band-limited so there
is no content above or equal to half the sampling frequency.
Bandwidth limiting is critical to prevent ghost frequencies
Computers, and other digital devices, count and manipulate
numbers using the binary system. The binary system works just
like the more familiar decimal system except for the number
base. The decimal system is base 10 and has ten symbols for
integers: zero through 9. When you run out of symbols, you
add a column at the left to indicate how many times you have
passed the base number and start counting again. Counting
goes like this: 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, (out of symbols, add a
column and start over), 10. Counting is in groups of 10 and
powers of 10. Each column represents a power of 10. The decimal number 11110 (base 10) equals 1x10 2, plus 1x10 1, plus 1
(10 0), which adds up to one hundred eleven.
The binary system is base 2 and has only two symbols for integers: zero and 1. Counting goes like this: 0,1, (out of symbols,
add a column and start over), 10. The binary number 10 is
equal to the decimal number 2. (In binary, I have 10 thumbs but
there is still only one on each hand.) Counting in binary is in
groups of 2 and powers of 2. The binary number 111 2 equals
1x2 2, plus 1x2 1, plus 1. Add these up and you’ll get 7 which is
the decimal equivalent of 111 in binary. The binary number system (base 2) can express any quantity just like the decimal
number system (base 10). The binary number will probably
have more digits but each digit will be either a zero or a one.
Actual quantities can be represented by different symbols in
different number systems. I have four fingers and a thumb on
each hand. All the fingers (4) plus the thumb (1) on my left
hand, combined with two of the fingers on my right hand (2)
add up to a number usually referred to as 7 (in base 10). The
binary number 111 is an alternative representation for the same
quantity. The binary number has more numerical digits but it
represents the same number of biological digits (fingers and
The binary system can be used for mathematical calculations
just like the decimal system. You can add, subtract, multiply
and divide in binary. The binary system is an ideal number system for machines because there are only two integer symbols.
These two symbols can be represented by a mechanical or
electronic switch with two states: open and closed. That’s how
computers work.
Essentially, a computer is a huge collection of switches. At an
instant in time, each switch is either open or closed, off or on.
With a sufficient number of switches, and very accurate timing,
you can calculate the trajectory to the moon and back, or
record music. With sufficient bandwidth you can transfer your
calculations, or your music recording, from one computer to
another by wire or microwave. Digital codes are very efficient
and convenient.
In the audio world, digital codes based on the binary system
are used to transform analog audio signals into digital data that
can be easily stored on optical discs. These optical discs are
the most popular storage medium for our high-end audio systems and they come in several flavors—CD, SACD, DVDVideo, DVD-Audio—each with special capabilities and characteristics.
Recording Music Digitally
are made
just like
with an
analog-todigital conversion, which converts an analog signal into a digital code. Digital-to-analog conversion, which converts that digital code back into an analog signal, must be performed during
playback. To help us choose desirable formats and hardware
we need to know a little bit about how the music was converted
into digital and about how it will be converted back to analog in
our home audio systems.
There are two digital coding methods commonly used for highend music reproduction: LPCM and DSD. This article will
attempt to shed some light on these. Compression algorithms
can be used to shrink LPCM data to fit in less space on the
disc or take up less bandwidth for transmission. Compressed
formats with lower fidelity (data reduced or lossy compression),
which are adequate for movie soundtracks and computer music
files, will be discussed briefly.
Multibit and 1-bit converters
The amplitude of an analog waveform can be represented digitally by multiple bits at relatively low sample rates (a little more
than twice the highest frequency to be stored) or by just a few
bits, or even a single bit, at relatively high sample rates (many
times the highest frequency to be recorded).
The term “bit” comes from combining the words binary and
digit. Sample rate is the number of samples taken each second. Data rate is the total number of bits per second. The data
rate equals the number of samples per second times the number of bits in each sample times the number of channels.
Compact discs (CDs) use 16-bit samples to represent signal
amplitude levels. Turning on various combinations of these bits
provides a representation of the amplitude of the analog signal
level at a specific instant in time. Sequential digital samples
can recreate the analog waveform. Super Audio Compact Discs
(SACDs) use many more samples with just 1 bit per sample to
recreate the analog waveform. Here are analogies to illustrate
how multibit and 1-bit systems work.
You could make a volume control from 16 switches and 16
resistors of different values. Switch number 1 (and its associated resistor) would allow the smallest signal level to pass.
Switch number 2 would allow twice the level of switch number
1 to pass. Switch number 3 would pass twice the level of
switch number 2, and so forth. The output from all switches
would be summed (added together). Each individual switch
would allow different amounts of the signal to pass and combinations of switches would allow the signal level to vary over a
wide range from completely off to full level with many steps in
between. These switches, like the bits in a CD sample, could
be turned on or off in various combinations to produce 2 16, or
65,536 different volume levels. 1-bit systems use a dif ferent
One switch could turn the volume all the way up or completely
off. Volume level could be controlled by turning the single
switch on momentarily, many times. If this switch were rapidly
turned on and off, with equal time in each position, the average
volume would be about half of maximum. Increasing the “on”
time would raise the volume level. Increasing the “off” time
would lower the volume level. The percentage of time that the
switch is “on” is called the duty cycle.
The on or off time could be increased by leaving the switch in
the on or off position for a slightly longer period of time. This is
analogous to pulse width modulation. Pulse density modulation
is similar.
In pulse density modulation, the frequency of pulses or the
width of these pulses is irrelevant. Only the average value of
the output—the percentage of time that the output is “high”—is
significant. If you want to learn how pulse density modulation
actually works, read the article Care and Feeding of the One
Bit Digital to Analog Conve rt er by Jim Thompson from the
University of Washington. It’s available here:
Linear pulse code modulation (LPCM or just PCM) is a coding
method using discrete samples that define the amplitude of the
analog signal at an instant in time. The analog signal is sampled at intervals and the sample amplitude is expressed as a
numerical value. The signal must be sampled at a little more
than twice the rate of the highest frequency to be recorded and
the size of each sample determines the resolution and dynamic
range of the signal.
LPCM coding is used to make regular compact discs with a
sample rate of 44.1kHz and a sample size of 16 bits. That
means that a sample is taken 44,100 times each second and
that each sample contains 16 bits which describe the amplitude
of the analog signal.
LPCM tracks on DVD-Video discs use a slightly higher sample
rate of 48kHz and larger sample sizes up to 24 bits. LPCM
coding with even higher sample rates can be used to make
DVD-Video standard discs—if little or no full-motion video is
included—and DVD-Audio standard discs. Sample rates up to
192kHz and sample sizes up to 24 bits can be supported on
DVD-Audio discs.
The sample rate establishes the bandwidth of the system
because the sampling theorem states that there can be no signal content at or above half the sample rate. To achieve bandwidth of 20kHz (a bare minimum for high fidelity audio) the
sample rate must be twice the highest frequency to be recorded with some headroom added to allow for filter attenuation.
The sample rate for a standard compact disc is 44.1kHz. This
permits a high frequency limit of 20kHz but requires a steep
filter (brick wall) to block frequencies just above 20kHz.
Higher sample rates of 48kHz or 96kHz or even 192kHz can be
accommodated on DVD discs because of greater storage
capacity and higher data transfer rate capability. These higher
sample rates extend the range of high frequencies and allow
the use of filters with improved transient response.
The compact disc sample size of 16 bits allows for 2 16 or
65,536 discrete levels of amplitude for a theoretical dynamic
range of 96dB. Some DVD disc standards allow samples up to
24 bits for 2 24 or 16,777,216 discrete levels. Samples
approaching 24-bit resolution can provide theoretical dynamic
range of 144dB, which exceeds what can be achieved with current electronic components due to intrinsic circuit noise.
LPCM coding is accomplished by analog-to-digital converters.
Modern converters with 20-bit resolution (or higher) use oversampling delta-sigma (also called sigma-delta) modulation to
raise sample rates and reduce sample size. Oversampling at
many times the required rate allows the sample size to be
reduced to as little as 1 bit.
The oversampled sigma-delta modulated bitstream is then
chopped up into discrete samples by decimation. Decimation
filtering, using a finite impulse response (FIR) low-pass filter,
quantizes the samples reducing the sample rate and increasing
the sample size for storage on the disc. This step is omitted
with DSD coding.
It’s important to remember that LPCM data on the disc is in the
form of discrete samples, even if that data was created using
sigma-delta modulation. It’s also important to understand that
the samples on a CD are 16 bits in size and there are 44,100
of them every second, regardless of the signal resolution during mastering. Oversampling or “upsampling” during playback
does not increase resolution or add to the information that is
recorded on the disc.
Resolution can only be increased by actually recording more
data, which requires more bits. LPCM with 24-bit samples at
96kHz can capture a lot more information than the CD standard
of 16 bits at 44.1kHz. So can DSD sampled at 2.8MHz.
DSD is a 1-bit digital coding process used to make SACDs.
The DSD bitstream is similar to an LPCM bitstream before the
decimation process since all modern LPCM recordings are
made with delta-sigma ADCs and delta-sigma ADCs are pulse
density modulators (see the above referenced article by Jim
Thompson). DSD offers much higher resolution than 16/44
LPCM and could be compared to 24/96 LPCM but DSD has a
simpler process.
DSD is a 1-bit digital stream at 2.8MHz. This digital bitstream is
recorded on disc essentially “as is.” The decimation step, used
to “downsample” the delta-sigma modulated bitstream into discrete samples for LPCM, is omitted. During playback the analog waveform is reconstructed directly from the DSD bitstream.
LPCM can also be reconstructed by sigma-delta modulation
and low-pass filtering but there is a significant dif ference.
LPCM data has been decimated into discrete samples for storage on the disc and the DSD bitstream has not. The
significance of this difference has often been underestimated.
Compression Algorithms
Even though DVD discs have a relatively huge storage capability, there still is not enough room for everything that content
providers want to put on the discs. Full-motion video and multichannel surround sound, alternative soundtrack languages and
a choice of audio mixes or compression types are some of the
things they want to include to make their products attractive to
a wider audience. Sound quality seems to be a minor concern
in many cases.
Getting all this stuff on the disc requires data compression.
Compression can be “lossless,” which suggests bit-for-bit
recovery, or lossy (data reduced), which means that some data
is discarded in the compression process.
Meridian Lossless Packing (MLP™) is a “lossless” compression
scheme used to make multichannel DVD-Audio discs. MLP™ is
claimed to provide bit-for-bit recovery with no loss of fidelity. It
may work as advertised and it may provide high performance
capability but we cannot verify that at this time. The DVD-Audio
players we have auditioned have offered poor audio performance by high-end standards. This poor audio performance may
be attributable to the players which have come from mass-market manufacturers, or to the recordings which have consisted
largely of remastered versions of older material, or to the mixes
which have consisted largely of ludicrous surround sound carnival tricks. We must reserve a final judgment of MLP™ for a
later date when we have had more experience with a wider
variety of software and hardware.
Dolby® Digital (AC-3) and DTS are lossy (data reduced) compression schemes used for multichannel soundtracks on DVDVideo discs. MP-3 is a lossy compression scheme used for
portable music devices like iPods and for computer storage of
music files. While these lossy compression schemes are perfectly acceptable for the purposes for which they were intended, they cannot provide high fidelity sound and are not suitable
for high-end music reproduction.
MP-3 should stay on your computer or portable device. Dolby
Digital and DTS can be enjoyed on a high-end audio system if
you have a video presentation to distract your attention from
sound quality but don’t consider either as a music source for
critical listening. Here’s why.
Compact disc is a marginally acceptable source for high fidelity
music reproduction. The data rate for CD is 44,100(sample
rate) x 16(sample size) x 2(number of channels) which equals
1,411,200 bits per second. If nine-tenths of this data is discarded using a lossy compression algorithm, the chances are very
slim that the remaining one-tenth could provide high fidelity
The “near-CD-quality” data rate for MP3 is 192kbps. Most
MP3s offer data rates lower than 192kbps. MP3 files provide
two channels of audio. Obviously, MP3 files are not an acceptable source for high fidelity music reproduction but what about
Dolby Digital and DTS?
Dolby Digital (AC-3) provides up to 5.1 channels of audio at
data rates that start below 400kbps. That’s a lower data rate for
each channel than MP3! You don’t have to be a rocket scientist
to figure out that these lossy compression formats can’t provide
high fidelity performance regardless of the quality of the hardware used for reconstruction. You don’t have to be any kind of
scientist, just listen.
DTS offers slightly higher data rates than Dolby Digital but the
data includes lots of side-band (not signal related) information
used to make the process “forward adaptive.” DTS sounds different from Dolby Digital but I don’t think the sound is necessarily better. It’s certainly no more suitable for our purposes
than Dolby Digital. Which would you rather have, a broken arm
or a broken leg?
Dolby Digital and DTS work fine for movie soundtracks and
music videos where your primary attention is on the picture.
Digital-to-Analog Conversion
The digital playback component in our home audio systems
must reconstruct an analog signal from the digital code on the
storage medium. This reconstruction is critical to the sound
quality we hear.
DSD material is a 1-bit pulse stream at a high frequency
(2.8MHz). Sony calls it “pulse density modulation,” which is
another way to describe delta-sigma modulation.
LPCM is a multibit code at lower sample rates of 44.1k to 192k.
LPCM can be converted to a 1-bit pulse stream at high frequency before digital-to-analog conversion by delta-sigma modulation.
Linear, multibit DACs can be utilized to decode the LPCM samples directly, or an oversampling interpreter and delta-sigma
modulator can be utilized to convert the LPCM samples to a
pulse stream at a higher frequency. Delta-sigma DACs include
a modulator which raises the sample frequency and reduces
the sample size.
A delta-sigma modulator is a circuit that translates a binary
number into a pulse stream with a duty cycle proportional to
the binary input. The duty cycle of a digital circuit is the portion
of time during which the signal is high. The pulse stream is
converted to an analog signal by averaging over time with a
low-pass filter. A delta-sigma modulator is a pulse density modulator.
A delta-sigma (or sigma-delta) DAC for LPCM usually combines
oversampling, delta-sigma modulation and filtering into a single
component. Delta-sigma DACs used to be 1-bit DACs and,
while there are theoretical advantages in linearity, 1-bit DACs
tended to deliver sound that was subjectively softer and less
detailed than multibit linear DACs (when decoding 16⁄44
LPCM). Multibit delta-sigma DACs have changed this conception. Today’s best multibit delta-sigma DACs can provide sound
that seems to equal the resolution of multibit linear DACs.
A DSD pulse stream can be converted to analog without the
delta-sigma modulation step because the DSD stream wasn’t
decimated for storage on the disc.
Optical Disc Players
Oversampling and Upsampling
failed to see through these smoke screens.
Oversampling is a term usually associated with conversion
from analog-to-digital or digital-to-analog. All CD players provide oversampling at even multiples of the sample rate.
Upsampling is a term usually associated with a digital-to-digital
process that may create sample rates that are not even multiples of the base rate—upsampling 44.1kHz to 96kHz for example.
No one seems to be able to offer a complete definition of
“upsampling” but it doesn’t seem to matter. Upsampling and
oversampling are marketing terms with little relevance to consumers.
Both processes, oversampling and upsampling, are mathematically equivalent and both affect the implementation of digital
filtering. Cascading digital filters may change the sound and
this may provide a subjective improvement but it does not provide higher resolution or add information to the recording.
Upsampling chips are readily available for a few dollars and
we’ve heard some players that use them to good effect but
upsampling is not a “magic bullet” that will turn your CD collection into high resolution digital. The recording process determines resolution. Tricks during playback can’t add (real) information that wasn’t recorded in the first place.
The High Fidelity Approach and Digital Audio
The high fidelity approach allows us to narrow the field of available products by eliminating from consideration those components that are poorly or inadequately designed. Learning a little
about how products actually work also helps us to avoid overpriced components that are made from the same of f-the-shelf
parts as well-designed products costing less. While the high
fidelity approach is easily applied to analog components, digital
audio components present some special obstacles. Many of
these obstacles are based on misinformation.
Digital audio is complicated and confusing to many consumers.
Some manufacturers have taken advantage of this fact by creating misleading advertising copy describing how their products
work. They have resorted to an old advertising dictum that
states “if you can’t convince them with facts, baffle them with
bullshit.” Some of the journalists writing about the subject have
Digital audio components must be designed in accordance with
basic scientific principles or they simply won’t work. Small companies can assemble digital audio components from integrated
circuits designed by qualified engineers at Texas Instruments or
Cirrus Logic. There are good and bad aspects to this situation.
The bad side is that it may be more difficult to narrow the field
of contenders without actually listening to a lot of components.
The good side is that even the poorest performing products
aren’t that bad. So how do we separate the products that are
worthy of our auditioning time from those that are not?
There don’t seem to be any easy answers to that question. We
will tell you what parts we find inside the digital audio products
we review, which will help you to determine some parts that are
common to the good or bad sounding players or other digital
components. Beyond that you’ll have to rely more on subjective
listening impressions from reviewers you trust or listen to more
products than you otherwise might.
The articles that follow will offer our listening impressions of a
variety of players and we’ll tell you how we arrived at our conclusions. The rest is up to you. APJ
ick has asked me to
say some things to
Journal readers about disc
players—to include my recommendations on which
players offer the highest
level of audio playback
quality and which players
should be avoided. I’ve
watched, listened to and
reviewed nearly 100 disc players over the last three years and,
believe me, I have plenty to say on this subject. Many of the
players I’ve reviewed have DVD-Video, redbook CD, DVD-Audio
and/or SACD playback capability. Let’s start there.
Most DVD players by the major manufacturers costing $250 or
more offer video playback performance that will satisfy all but
the pickiest people with the biggest, highest resolution displays.
Optical Disc Players
I use $250 as a jump-off point because that’s the price of
Sony’s DVP-NS755, a progressive scan DVD player with 12bit/108MHz video DACs and multichannel SACD playback.
That player is good enough that very few RPTVs have the kind
of resolving power to differentiate between it and the players
that better its video quality. But the audio quality is a dif ferent
story entirely.
A lot of people I know want a DVD player that’s also a reference quality CD player—a single box that performs two critical
functions. I’m speaking of reference quality in audiophile terms,
not in generic home theater magazine terms. Many people writing for the mainstream AV press have never heard a true reference quality CD playback system and can typically identify only
the stuff that’s better than the crap they’ve heard. I was in this
camp until I started spending time at Dick’s place during my
tenure at Widescreen Review.
The sad truth is that, at least in audiophile terms, I’ve heard
only two DVD players that are true reference quality CD players suitable for use in high-end audio systems: the Ayre D-1x,
and Arcam’s FMJ DV27. The simple explanation for this is that
it isn’t easy or inexpensive to build a good dedicated CD player. And DVD players are inherently different and require a lot
more circuitry, costing more to build.
More elaborate drives are required, as are laser pickup mechanisms that can retrieve both DVD and CD standard data. Also
needed are MPEG decoders and video DACs in addition to the
audio DACs, and in many cases 5.1-channel Dolby Digital
and/or DTS decoders along with six channels worth of DACs
and analog outputs. On top of that, DVD-Video standards dictate that PCM and Dolby Digital signals on DVDs are at a sample rate of 48kHz, not the 44.1kHz of redbook CD. The better
players need to operate at separate clock speeds too, one for
each sample rate. That doesn’t leave a lot of money for trick
power supplies or high quality analog output stages. The bottom line is that there’s more stuff in the box, and it costs more
to make a box that excels at more than one thing. And DVD
players are primarily video products, and that’s where the priorities typically lie.
We’ll look at the two DVD players that are at home in any highend audio system as CD players, and a third that’s a terrific
SACD player. That brings up the question of what makes a
good player.
What Makes A Good Player?
In his series on digital controllers for Widescreen Review, Dick
pointed out that many controllers use similar parts—surround
sound decoding/processing chips, DSPs, DACs, etc.—in spite
of the wide disparity in prices among many of the products.
Why pay more for a given box if it has the same parts inside
that a competitor offers for half as much? Disc players aren’t
quite as cut and dried in this regard as controllers often can be,
but there are some things I’ve observed while performing these
With digital, the first things I look at and think about are the
DACs. You’re never going to get any more resolution out of a
disc player than what comes out of the digital-to-analog conversion. The best you can do after that is mitigate any further signal degradation, period. Of the few conclusions I can draw, one
is of the inherent quality of the Burr Brown PCM-1704 DAC.
The best digital controllers I’ve heard use this DAC, including
the Theta Casablanca II Extreme DAC version, CAL’s CL-2500,
and Integra Research’s RDC-7. Several of the best performing
disc players I’ve heard also use it, including Denon’s DVM3700 (the first DVD player I heard that was also a respectable
CD player), Ayre’s D-1x, Wadia’s 861, and Resolution Audio’s
Opus 21. Every product I’ve heard that uses this DAC competes with the best products in its category, or establishes the
benchmark for a particular category. Odds are in favor of good
sound when this component is used.
While many of the initial 1-bit sigma-delta DACs offered soft, illdefined sound quality in the controllers and disc players that
used them, the new multibit sigma-delta designs are a huge
step up in performance. Dick first noticed this with the Sunfire
Theater Grand II controller, which used Analog Devices’
AD1853 DACs. We’re hearing terrific sound from other products with multibit sigma-delta DACs sourced from Wolfson and
Burr Brown, among others. Although many factors are certainly
at work, these DACs seem to represent a huge step-up from 1bit designs with respect to overall resolution of detail, focus,
and soundstaging. Players that we’ve heard that use them
include Arcam’s DV 88 and FMJ DV27 players (DACs by
Wolfson), Ayre’s CX-7, and the Musical Fidelity Tri-Vista (both
use the Burr Brown PCM-1738).
The track record for Cirrus Logic’s sigma-delta DACs is much
more enigmatic. Pioneer’s DV-47A uses the Cirrus Logic
CS4392, and both Classé’s Omega SACD-1 and Philips’
Optical Disc Players
SACD-1000 use the Cirrus Logic multibit sigma-delta CS4397.
All offer disappointing audio performance. On the other hand,
Philips’ DVD-962SA also uses the Cirrus Logic CS4397 and
Simaudio’s Moon Stellar uses the Cirrus CS43122, a 5-bit
sigma-delta stereo DAC with a specified 122dB dynamic range.
Both sound terrific. The Simaudio features a massively overbuilt power supply and balanced circuitry in the analog output
$1,000 don’t offer any of the above accoutrements. More often
than not, these players will have cheap, poor quality DACs,
switching power supplies, and poorly implemented analog outputs stages. Caveat emptor. You’re more likely to get great
video performance and mediocre audio performance from most
products in this price category.
The power supply and analog output circuitry are other areas
that have enormous impact on a player’s sound quality, but
exactly how that’s accomplished is more mysterious. Some of
the Sony players (the $1,500 DVP-9000ES and $3,000 XA777ES for example) are built more robustly in both regards and
sound much better than the other players in their price categories and aren’t embarrassed at all by comparison to products
that cost many multiples of their prices. Here are a few other
things I’ve observed that may help you make better decisions.
I attended an Audio Perfectionist seminar and asked Dick
what attendees of his other seminars were especially concerned about. One thing that became clear to us is that many
enthusiasts lack an appropriate strategy for building a system
that’s musically satisfying. That’s easy to understand; there’s a
lot of misinformation out there to sort through. Although Dick
and I will discuss system building strategies in future issues of
the Journal, I want to offer a couple of quick tips here.
Other Buying Tips
Linear power supplies with larger (and/or multiple) transformers
usually sound better than inexpensive switching power supplies. Sony’s DVP-9000ES uses a linear supply with dual transformers. Arcam’s FMJ DV27 and DV-88P players are virtually
identical (including the DACs) with a couple of exceptions—the
FMJ player uses an Accousteel chassis and has separate
power supplies (and transformers) for analog and digital circuitry. The differences in audio performance between these two
players are not subtle—the FMJ player sounds significantly better.
With the analog output stages, the answers are far less distinct. Ayre’s D-1x shows how killer discrete circuits can be, but
Ayre’s CX-7 also shows just how close well designed ICs can
come to that level of performance. Sony’s DVP-9000ES
demonstrates that off-the-shelf op amps aren’t necessarily a
death sentence for terrific sound either.
When you see XLR connectors on the back of the player make
sure to find out, from either your dealer or another source,
whether the player is truly balanced. Some products aren’t truly
balanced, but simply use a phase splitter in front of XLR connectors to create “balanced” connections. This can sometimes
result in worse sound quality from the “balanced” outputs than
the single-ended outputs.
Typically, DVD/CD/SACD/DVD-A players selling for under
If picture quality is your priority, many DVD players available for
under $1,000 offer good enough video quality for most displays. If you’re willing to spend more on a player and display,
and if you’re also a hardcore videophile, the higher priced components may give you more of what you need. If you’re more
budget conscious, focusing your money elsewhere in your system will yield higher quality audio performance. Also, the CD’s
days are numbered. High resolution digital in the form of SACD
and DVD-A are here, and at least one of these is likely to survive as a high-end niche format. Think before spending
megabucks on a CD playback rig right now. Even if I had that
kind of jack (which I don’t) I personally wouldn’t be considering
spending $20K-$30K on a CD rig that doesn’t also play back
SACD and/or DVD-A, or at least offer an upgrade path to either
or both. (You’ll read about a $20K dCS CD/SACD rig here that I
would spend the money on, if only I could.)
As you’ll read below, $3,000-$3,500 can buy you a hell of a CD
player. Better performance certainly is available for more
money, and we’ve identified some of the products you should
look at. But I think those of you on limited budgets would do
better with one of these excellent $3K CD players, investing
more money in your speakers, amps, preamps, etc. For an
investment of $4K—the cost of a good CD player and $500$700 for a used Sony DVP-9000ES—you can get surprisingly
close to the best CD/DVD/SACD performance available at any
price. Then, after the rest of your system is dialed in, you can
look into upgrading these source components as necessary.
Optical Disc Players/Ayre D-1x
The Recommended CD/DVD-Video Players
In addition to playing DVD-Video discs another advantage of
nearly all recent vintage DVD-Video players is that they will
play back DVD-Video standard Audio Discs (DADs) from
Classic Records and others. These discs have uncompressed
two-channel PCM tracks at high resolution (24/96 in most
cases). Although there are relatively few of these discs on the
market, those I’ve heard sound terrific. Some of the audiophile
labels, like Chesky and AIX, will be including high resolution,
uncompressed two-channel PCM tracks on their DVD-A discs
and it may be a significant benefit to you to be able to play
these tracks, which sound noticeably superior to redbook CDs.
As I mentioned, I’ve only used two DVD players that are also
reference quality CD players in the strictest sense. Note that,
although I do have experience with a massive number of players, I’ve not yet heard the Meridian players, which also feature
DVD-A playback, nor have I heard some of the other high-end
jobs from companies like Krell and EAD. This isn’t meant to be
a totally comprehensive survey, but a summary of my experiences and recommendations of the players I have heard.
Ayre D-1x DVD Player
Ayre’s D-1x sells for $8,000 with CD playback capability and no
video, and for $11,500 fully loaded with audio and progressive
scan video. The D-1x uses an outboard power supply with AC
line filtering and separate choke-input filters to further eliminate
power supply-related noise and artifacts. It eschews the use of
op-amps for discrete circuits.
Two K-Grade 24-bit linear Burr-Brown PCM-1704 DACs are
used in a dual-differential configuration for each channel of the
D-1x. Burr-Brown’s DF-1704 filter is used, which performs 8x
oversampling with signals up to 96kHz. Since it is a DVD player, it will play back DVD-Video standard audio discs (so-called
DADs) with high sample rates so long as the tracks are uncompressed PCM, not MLP-based DVD-Audio standard discs. The
parts for this dual DAC combo cost nearly $46 per channel plus
$10 for the filter, which totals just over $100 in parts costs for
DACs and filtering alone. That doesn’t sound like much, until
you consider that many CD and DVD players use sigma-delta
stereo DACs with integrated filters that cost $3 a piece, or even
less (perhaps much less in mass quantities).
There are two master audio clocks for 44.1k and 48k signals,
when one is engaged the other is turned off. At higher sample
rates the 44.1k clock is used for 88.2k signals and the 48k
clock is used for 96k signals.
Fully balanced circuitry is used, and the D-1x sounds best from
its balanced outputs. As with all of A yre’s gear, no feedback is
employed, and the chassis is non-ferrous aluminum which
resists magnetic fields and eddy currents. Few players are in
the Ayre’s league with respect to construction and parts quality.
If someone tries to sell you a player that’s more expensive than
the D-1x, beware that you’re not likely buying superior parts, or
construction, and therefore you’re not buying more performance.
Charlie Hansen, Ayre’s president and chief designer, believes
this to be the purer approach, although much more expensive.
When I interviewed Charlie, he analogized using discrete circuits with baking a cake—would you rather eat a cake that’s
made from scratch by a French pastry chef, or would you
believe that Betty Crocker’s off-the-shelf cake mix would taste
just as good?
For those who care, the D-1x tells an equally compelling story
on the video side. Over three years ago when it was first introduced, Ayre’s player had tricked-out (and expensive) 14-bit
video DACs that sampled at 54MHz stacked in a dif ferential
configuration. (I’m still not aware of another player that uses
dual-differential video DACs.) The D-1x uses Silicon Image’s
(formerly DVDO) progressive scan chip set for de-interlacing
and 3/2 pulldown detection and compensation. With film-based
material this chip set is second to none, and only Faroudja chip
sets and the very latest chip sets used by Sony better its performance with video-based material. (Most concert videos on
DVD, for example, are video-based.)
The only downside to the D-1x’s video prowess is that, as great
as it is, it’s a 480p player. People with the kinds of systems
Optical Disc Players/Ayre D-1x
required to see how great this player really is (i.e., expensive
front projection systems) need the video scaled up to resolutions higher than 480p for optimum playback. A yre’s recent
announcement of an SDI (Serial Digital Interface) output for the
D-1x is a positive step. That will allow the D-1x to be connected to high-end video processors entirely in the digital domain,
and may make the D-1x even more attractive to the high-end
video crowd.
In my opinion, the D-1x is quite simply the best CD/DVD player
available in terms of pure performance. I’ve heard nearly all of
Ayre’s products, and the first thing that jumps out is the complete absence of noise and distortion. Such is the case with the
D-1x. The music is presented from a pure black background,
and words like clarity and purity come immediately to mind.
The audio quality is detailed, revealing nuances of your best
and most familiar CDs that you’ve not heard before. There are
layers of space and detail, front to back, with great spatial separation of musicians on stage. Image focus is razor sharp with
amazing delineation. Sound permeates far beyond the speaker
boundaries—the size of the soundstage will be determined by
the rest of your system and your room, not the D-1x. Tonally, its
performance is beyond reproach, with weight and authority at
the low end, airy extension at the high end, and precision and
focus at all frequencies in between.
The D-1x is so good that its only fault may be that it’s just too
good for the compact disc! It ruthlessly reveals the shortcoming
of the format, especially when you put on a 24/96 DAD from
Classic Records and hear how much better it sounds than redbook. As long as the rest of your system is up to the D-1x’s
standard, you’ll get layers of spatially precise images front to
back, and way out to the sides of the soundstage. But along
with that resolving power comes a sound that can be a bit lean,
and a little on the cold side.
Ayre’s two players (the new CX-7 is covered later in this issue)
seem to ratchet the music upward in frequency just a bit, lacking some warmth and therefore spotlighting the midrange and
treble to a degree. Using cables with a warmer midrange will
ameliorate this somewhat, but other deleterious side ef fects
may accompany that desirable characteristic.
power but images in a softer, rounder fashion and sounds more
musical, more full-bodied and enjoyable to listen to in non-hi-fi
terms. The Wadia, however, doesn’t do DVD-Video, nor can it
play back 24/96 DADs without using a compatible player as an
external transport.
But listening to the music swell and become fuller and richer
with the 24/96 material, while maintaining the same mind-boggling resolution, I came to think that the D-1x simply delivers
everything the CD has, ugly warts and all. I personally hope
that there’s a high resolution player (SACD and/or DVD-A) in
Ayre’s future—a player of the D-1x’s level could establish a
new benchmark for high res players. On the other hand, the D1x probably wouldn’t be a good match for a system built
around speakers tipped toward any degree of brightness. And
note that my evaluation here is based upon the “listen” lowpass filter setting; the “measure” setting is noticeably less
musical in comparison. Listen was the one that felt right to me
and cemented in my mind the D-1x’s place at the very top of
available CD players. I’ve heard players that evoked a more
favorable emotional response (including A yre’s own CX-7,
which you’ll read about later) but I’ve never heard one I could
unequivocally say was better in terms of absolute quality.
With 480p DVD-Video the D-1x is just as resolute. Every detail
is revealed, which imparts as much depth and dimensionality
as I’ve seen from a progressive DVD player. The images take
on that rarified illusion of looking into real three-dimensional
space. If 480p is enough to drive your display, the D-1x is the
best you’ll find, although I must add that recent players, like
Sony’s DVP-NS999ES and Pioneer Elite’s DV-47A, are more
than good enough for most displays for a mere fraction of the
price. Spending just $1,100 on one of those players would
allow you to put a lot of money into the rest of your system.
Disc access and transport speed are not equivalent to the Sony
and Denon players. But if you’ve got the money to spend, the
D-1x is the undisputed heavyweight champion of pure performance CD/DVD players.
By comparison, Wadia’s 861 shows nearly the same resolving
Ayre Acoustics
2300-B Central Ave.
Boulder, CO 80301
Optical Disc Players/Arcam FMJ DV27
Arcam FMJ DV27
The Arcam FMJ DV27 is just $2,599 and performs so well as a
CD/DVD player that it will be difficult for many people to justify
spending any more.
Technical refinements include Wolfson WM8740 multibit sigmadelta DACs, separate power supplies for analog and digital circuitry, a disc drive suspended on four semirigid tiers and a “Full
Metal Jacket” chassis with two layers of Acousteel surrounding
a layer of rubber to damp vibrations.
using high-end front projection rigs will likely never see anything to be concerned about from the DV 27. Its progressive
scan solution is provided by Silicon Image and is excellent.
The DV27 is bested in audio quality by Arcam’s own FMJ CD23
and some of the other (much) more expensive dedicated CD
players in this survey. What sets it apart is its complete package of high performance with DVD-Video and CD at a price
that’s so much lower than Ayre’s D-1x, which is the only player
that outperforms it with both video and audio. At $2,599 it’s
nearly the best in both categories, and many people will be
able to make a case for spending the extra bucks in other
areas of their system. Also note that Arcam is working on a
DVD-A upgrade for the FMJ DV27. Existing players will be
upgraded in the field.
Three clock speeds are used: one for the video DACs and two
for audio DACs—one for 44.1kHz and its multiples (88.2k and
176.4k), and another for 48kHz and its multiples (96k and
US Distributor: Audiophile Systems Ltd.
8709 Castle Park Dr. Indianapolis, IN 46256
888 272 2658
The Arcam FMJ DV27 is simply wonderful. While it offers very
good overall resolution, its main strength is its exceptionally
natural sound quality, especially with vocals. It soundstages
and images right up there with the best of them, allowing sound
that’s not at all confined to the physical loudspeaker boundaries. It is slightly rough around the edges in the midrange, but
I found that to be an attractive quality. That little edge makes
for a visceral and immediate vocal presentation that I really get
Dick and I initially compared this player to the Wadia 861 and,
while the Wadia was better, it took a while to figure that out and
longer still to quantify the differences between the two players.
The Wadia is more refined, with better definition in the
midrange and extension through the treble, and a grander,
more precise spatial presentation. Note that all these are subtle
differences that will only be noticed in high resolution systems.
The Arcam’s video presentation is also phenomenal, only bested in fine detail by the Ayre D-1x and the most recent vintage
of mainstream flagship players like Sony’s DVP-NS999ES,
Pioneer Elite DV-47A, and Denon’s DVD-3800. These newer
players offer an extra dollop of resolution that will be noticed by
those with the highest resolution displays. Those who aren’t
DVD-Video Players That Also Play DVD-A/SACD
Universal players (those that play CD, DVD-A, SACD and
DVD-Video) are thus far a better idea than reality. In other
words, I haven’t heard one yet that offers more than mediocre
performance with all formats. The typical pattern I’ve seen is
decent performance with CD and DVD-A respective to the player’s price category, excellent video performance, and poor performance with SACD. Players in this category include Pioneer
Elite’s DV-47A, Marantz’s DV-8300, and Onkyo’s DV -SP800.
Multichannel just compounds these problems—it’s hard enough
to get high quality two-channel out of a player let alone six
channels with multiple formats (and that doesn’t even address
the issue that there aren’t any reference quality six-channel
preamps on the market). High quality DACs and analog output
stages are expensive, and the cost increases threefold with
multichannel. As an example, the Pioneer Elite DV -47A uses
three Crystal CS4392 DACs for PCM and DSD in order to hit
its $1,199 price tag. These integrated DAC/filters have an OEM
quantity cost of about $3 a piece. I’ve heard only one
DVD/SACD player I can unequivocally recommend, and no universal or DVD-A players. As a result, I own a Sony DVP9000ES SACD player but not a DVD-A player. Let’s discuss
some specifics.
Optical Disc Players
Recommended: Sony DVP-9000ES
Almost Recommended Players
Any discussion of this category of products has to start with
Sony’s DVP-9000ES. This DVD/SACD player was a steal at its
introductory price of $1,500. It’s overbuilt with a big power supply and tons of copper shielding, and slick good looks with a
slim-line disc tray. Its DACs are LSIs built by Sony. It offers
excellent performance with all formats except the CD, with
which it offers merely competent (soft but inviting) performance.
The following players are solid bargains and offer excellent
video performance, but aren’t quite as suited to dedicated highend audio installations as the Sony DVP-9000ES.
This player is now discontinued, but can be found on the internet for well under $1,000 and it’s tough to beat even by spending much more. The video performance is good enough that
only the most recent flagship-level players outperform it, and
even then not by a significant margin.
Its performance with two-channel SACD is so good that only
the dedicated audio players like the Sony XA-777ES and
Musical Fidelity Tri-Vista outperform it, and then only by a relatively slim margin. Although it is only respectable with CDs, it
does err on the soft and inoffensive side, not the harsh, edgy
side. And it sounds remarkably good with 24/96 DADs, suggesting that its sound quality increases commensurately with
the quality of the incoming signal. The DVP-9000ES is an
SACD player that will stand on its own in the highest resolution
systems. Richard Hardesty and Richard Vandersteen each also
own a DVP-9000ES, which says something.
For those who are curious, the new Sony flagship DVPNS999ES is a better video player, and a slightly inferior SACD
player. Although it offers multichannel SACD (compared to the
DVP-9000ES, which is two-channel SACD only) it is not the
overbuilt, over-achieving monster its predecessor was with twochannel SACD. At this point multichannel doesn’t interest me
nearly enough to want or recommend a new DVP-NS999ES
over a used DVP-9000ES.
I’ve always appreciated the audio quality of Denon DVD players starting way back with the DVM-3700 carousel player,
which was a good CD player. As far as the mass market players are concerned, Denon’s are always respectable. They typically offer as much resolution as any of them, if not more, but
also have the relaxed, non-mechanical sound of the Sony players. Their recent players, the DVD-9000 ($3,800) and DVD3800 ($1,199), are two of the very best video players available—both use 12-bit/108MHz video DACs and Silicon Image’s
vaunted SiL504 progressive scan solution.
Both are also very good with CD and DVD-A, and Denon players are right up there with Sony players on disc access speed
and ease of use. The DVD-9000 uses Burr Brown PCM-1704s
in a dual differential stack on the front channels, while the
DVD-3800 uses the new PCM-1738. In spite of its build quality
advantages, the DVD-9000 doesn’t offer enough of a performance edge over the DVD-3800 to justify the massive price differential. The DVD-3800 is a superb machine and does everything very well.
Philips DVD 962SA
Although I was not at all impressed with the Philips SACD
1000, that company’s first and most expensive DVD/SACD
player ($2,000 upon introduction), the DVD 962SA is much better at just $599 MSRP. It has Faroudja progressive scan, with
overall video performance that’s just short of the Denon DVD3800 and Sony DVP-NS999ES. More germane to this conversation, its SACD audio quality is right up there with the Sony
DVP-NS999ES, although short of the excellent performance of
the DVP-9000ES. A very solid double duty player; its clunky
user interface is the only detractor.
Players To Avoid
These players aren’t necessarily awful—though some of them
are—but they don’t represent good value in terms of performance and cost. These are not recommended for purchase but if
you already own one you probably won’t get cancer because of it.
Ayre CX-7
Classé Omega SACD-1
I had high hopes for the $12K Classé Omega SACD-1. It’s a
two-box solution with a separate power supply in one box and
a beautiful, top-loading player in the other box. The fit & finish
of the shipping container is nicer than most players!
It uses three Cirrus Logic CS4397 stereo DAC chips for PCM
and DSD—two chips stacked in a dual-differential configuration
drive the balanced outputs and a single stereo chip for the single-ended outputs. The sample I heard was awful. It made both
CD and SACD sound like some of the worst, most non-musical
digital I’ve heard. It sounded so bad that I wondered if something was wrong with the player.
I’ve never used a player that malfunctioned in such a way that
everything operated perfectly with bad sound being the only
symptom but I suppose anything is possible. In any case, if a
dealer auditions this player for you listen carefully to the player,
not the salesman’s accolades, and certainly don’t be swayed
by the fact that Stereophile includes this player in their annual
recommended components issue.
The CD performance is best with the Pioneer Elite player,
which uses Pioneer’s Legato Pro upsampling technology to eek
some extra air and detail out of PCM material making it a
respectable performer among DVD players in its $1,199 price
range. The DVD-A performance of all three players is nothing
to write home about, and SACDs sound mechanical and digital,
like merely decent redbook CD playback. Dick once said of an
early DVD/DVD-A player that it does everything it’s supposed
to do except sound good. That’s pretty much the story here.
Note that Pioneer has two more recent universal players that
use different DACs—the Elite DV-47Ai and the Pioneer DVD45A. I’ve not yet had a chance to audition these players, and
perhaps they are improved. APJ
Dedicated Audio Players
The following players are for music only and we’re
giving them more extensive coverage with complete
reviews. All are recommended and each has some
special strengths.
Philips SACD 1000
You’ll see this player at closeout prices of less than $1,000, but
don’t fall for it. It was overpriced at its original $2,000 price and
the new players smoke it in audio and video quality.
With video it doesn’t have progressive scan and frankly isn’t
very good. Like the Classé Omega player, it too uses the Cirrus
Logic CS4397 DACs. Its audio quality, like some of the players
below, is so poor that it makes SACD sound like mediocre CD
he Ayre CX-7 costs $3K and boasts a new industrial design
that matches Ayre’s latest line of products, like the K-5x
preamp. The chassis is aluminum and it’s a pretty look, albeit
spare. I admit to being a less-is-more kind of guy when it
comes to this kind of thing, so its aesthetics speak to my sensibilities in particular. The CX-7 is a single box with an integrated power supply and uses ICs as opposed to the discrete circuits of the D-1x.
Pioneer Elite DV-47A/Marantz DV-8300/Onkyo
All three of these universal players are based on Pioneer kits
and offer similar performance. The Pioneer 47A is $1,199. The
Marantz beefs up the power supply a bit and costs more at
$1,599, while the Onkyo costs $200 less at $1,000 MSRP.
The video for all three players is red hot and among the very
best available at any price although Pioneer’s proprietary deinterlacing solution is now slightly behind those offered by
Faroudja, Silicon Image, and even Sony’s latest generation of
video processing.
The digital front end of this player uses Burr Brown’s PCM1738 stereo DAC, which is billed as an “Advanced Segment”
DAC, and a Burr Brown DF-1706 digital filter. The DF-1706
employs 4x oversampling, and is cascaded with the PCM1738’s integral 8x filter.
Opus 21
According to Burr Brown, 16-bit signals are interpolated to 24
bits, which then are split, or “segmented,” into the upper six
bits of the signal and the lower 18 bits. The lower 18 bits are
converted in a multibit sigma-delta process, while the upper six
bits are converted in a linear conversion. This stereo DAC is
configured for differential (balanced) output for each channel,
and is a current output.
Ayre’s Charlie Hansen believes it’s a huge advantage to have a
current output DAC, which allows him to design zero feedback
current-to-voltage converters that are superior to the op amps
used within typical voltage output DACs. Whatever the reasons, the early returns show the PCM-1738 to be something of
a price/performance breakthrough. It costs just $5 (list price)
per stereo chip and, as noted, only one chip is required for a
balanced stereo output. That’s a fraction of the price for a dualdifferential PCM-1704 set-up, and the players that use the
PCM-1738 absolutely hold their own with the PCM-1704equipped players.
The CX-7’s analog output stage is truly balanced. The CX-7
also has user-adjustable low-pass filter settings of “listen” and
“measure,” and like the D-1x it sounds best in “listen” from the
balanced outputs.
Although I did not have both players on hand for A/B comparisons, the Ayre CX-7 gives up remarkably little ground to my
memory of its big brother, the D-1x. The CX-7’s sound is clean
and pure, with plenty of resolution. Frequency extension on top
seems roughly the same, but the bottom end of the CX-7 is a
little fuller, although perhaps just a bit looser and less defined,
resulting in a slightly (but decidedly) warmer overall character
than the D-1x. I reiterate that the CX-7 also shares that characteristic of shifting the music up in register a hair—it’s not exactly bright, but it’s certainly not reticent in the midrange/treble,
and the body of the music lightens just a hair.
The dynamic contrast is also staggering, and better than any
player I’ve heard other than the D-1x. It’s spatiality and focus
are terrific but somewhat flatter from front to back than the D1x or Wadia 861. The CX-7 places musicians and vocalists on
the stage with impressive solidity, although it’s slightly less
dimensional and holographic than the Wadia 861. Still, at $3K
the CX-7 is excellent in all respects and in some ways I find it
even more emotionally engaging than its big brother, the D-1x,
which too often reminds me just how flawed 16/44 CD really is.
If I were shopping for a CD player in the $3,000 range, the CX7 and the Arcam FMJ DV27 are the two I’d consider. The
Arcam doesn’t quite match the striking sense of clarity and
overall transparency of the CX-7, but it has more body, enough
detail, and it unquestionably makes music. And the FMJ is a
kick-ass progressive DVD-video player and will also play DVDvideo standard discs with uncompressed 24/96 tracks, with a
full DVD-A upgrade in the works.
These are both excellent players and are close enough in
audio performance that your preference for one over the other
is likely to be system and taste dependent. Personally, I could
live happily with either one. Audition both and see which one
fits your needs and sounds best in your system. APJ
Ayre Acoustics
2300-B Central Ave.
Boulder, CO 80301
www.ayre.com 303-442-7300
esolution Audio sells the Opus 21 direct over the internet
and through a few select distributors in Europe. Selling
direct allows Resolution Audio to offer this collection of engineering and high quality parts for a modest $3,500 (this price
reflects an anticipated increase).
The Opus 21 is a two-box system, with the power supply and
display electronics (i.e., the noisy stuff) separated from the
transport, DAC and analog output stages. Both boxes are
Opus 21
attractive, and the overall fit & finish is hot for a $3,500 player.
The chassis are both aluminum to eliminate magnetic interference. The Opus 21 is now shipping with an improved disc drive
that is said to be quieter than its predecessor. It has a digital
input that will accept external PCM signals up to 96kHz, and it
also has its own volume control, which allows users to drive an
amplifier directly from its analog outputs. Although the volume
is controlled digitally, the volume control is implemented in the
analog domain to avoid truncating any resolution.
The Opus 21 uses Burr Brown PCM-1704 24-bit DACs stacked
in a dual-differential (balanced) configuration. Although these
are the same DACs used in A yre’s D-1x, they aren’t the Kgrade variety. The K-grade DACs are guaranteed to meet certain measured performance specs (note that there is no guarantee that the ungraded DACs don’t meet the same specs). In
front of the DACs are Motorola DSP-based filters using software-based algorithms written by Resolution Audio that oversample at 16x with 44.1k and 48k signals, and at 8x with 96k
A passive current-to-voltage converter is used, followed by an
instrumentation amplifier to achieve the proper gain characteristics. Resolution Audio believes this approach offers superior
common mode rejection compared with discrete circuits,
although part of this tradeoff is that the instrumentation
amplifier employs feedback. All in all, you’d be hard-pressed to
find a player with a more unique design, or with as much
money in parts/construction anywhere near a $3,500 retail
The Opus 21 has knock-out resolution from top to bottom, with
excellent extension at both frequency extremes. The midrange
is not only well defined, but smooth in character. Low level
musical details are clearly audible. To a certain extent this resolution is a double-edged sword.
On the one hand, with good recordings it can be a striking
experience to hear so much more of what’s on the disc—especially with 24/96 material via the digital input. Most people have
never heard the kind of resolution this player delivers, let alone
at the $3,500 price point. I’ve heard much more expensive
players that aren’t as revealing.
The only criticism I can level at this player is that, as high in
resolution as it is, the musical picture doesn’t hang together in
as coherent a fashion as I’ve heard from other players. This is
a difficult phenomenon to explain. The best I can do is offer the
example of a guitar playing.
While I could certainly hear all the pieces of its sound with
great clarity—the strings being strummed, the hands sliding on
the strings and hitting the body of the guitar occasionally, and
the decay of the string’s sound—those individual elements
never coalesced into a coherent image of a guitarist playing his
Ayre’s CX-7 is in the Opus 21’s price range, and while that
player doesn’t have quite the same level of resolution in hi-fi
terms, it does present images of musicians (and the music
overall) as more of a whole, coherent picture that I found more
Consider a few things though. I use time- and phase-accurate
Vandersteen speakers. Time- and phase-accurate speakers
image in a more convincing, dimensional, and holographic
fashion than conventional designs and I’m convinced they also
reveal things about components that other speakers don’t. In
any case, I think that the $3,000-$3,500 price range is a killer
elbow of price/performance for enthusiasts.
High res digital is here, and if you’re an audiophile you’re going
to want that capability in the future. But for now, I think this is a
reasonable price range in which to shop for something that will
help you get the most out of your CD collection while the library
of high res titles grows. Although I didn’t respond emotionally to
this player as I did to both the Arcam FMJ DV27 and Ayre CX7, it is outstanding in many ways and many audiophiles will get
off on the staggering amount of musical detail they’re hearing
and forget the rest.
Resolution Audio makes it easy for you to decide for yourself—
they offer a 30-day in-home audition period. If you decide to
return the player they’ll charge you a 5% re-stocking fee to
cover the credit card transaction fees and shipping charges.
Check it out. APJ
Resolution Audio
415 643 6971
dCS Verdi & Delius
’ve been an audio enthusiast and hobbyist for years. To make
a living doing something that is also my hobby is immensely
enjoyable and satisfying, and I’m ever thankful for that. But, as
with any job, a sense of monotony can occasionally set in,
making things feel like a grind. Then a dream assignment like
reviewing this dCS gear drops into my lap and bam! I’m
charged with a sense of purpose and renewed (wild!) enthusiasm.
I imagine the anticipation I felt while waiting for the dCS gear to
arrive is akin to what the Car and Driver guys feel when
Ferrari’s latest super car is scheduled for delivery—Corvettes
are fun to drive, but they’re not quite up to the same level of
hedonistic thrill. Like the Ferrari, dCS’ gear is nothing short of
an attempt at defining state-of-the-art performance. I may
never find out what it’s like to fly around the race track in one
of those 12-cylinder land rockets, but those guys probably
aren’t going to get to listen to dCS’ Verdi/Delius combo either!
For the last two years at CES, the UK’s dCS has been making
spectacular digital sound in the various rooms occupied by
Audiophile Systems Ltd., their US distributor. Among the products I’ve heard making killer sound at the shows are the Purcell
digital-to-digital upsampler, which is said to convert 16/44 CDs
to a 1-bit/2.8MHz DSD stream, and the Elgar Plus DAC. The
depth and dimensionality this combo was extracting from CDs
was truly remarkable. With this in mind I literally jumped at the
chance to review dCS’s Verdi CD/SACD transport and Delius
DAC (little brother to the Elgar Plus).
While dCS’s flagship playback system is the $34K
Verdi/Purcell/Elgar Plus (yep, that’s $34K with three zeroes—
why do you think I went with the Ferrari references?), the
Verdi/Delius combo comes in at just $20,990 ($10,995 for the
Verdi and $9995 for the Delius). That’s not chump change, but
given that Linn, Levinson, Krell, and others sell CD-only players
for as much or more it’s not out-of-this-world either. Audiophile
Systems considers the Verdi/Delius combo a statement of functionality and performance at this price point, and I agree.
Before I get to all that though, just who is dCS?
dCS: From the Military to Your Listening Room
dCS takes great pride in regarding most manufacturers of highend digital playback devices more or less as “assemblers.” The
distinction they offer is that, where most manufacturers source
their DACs, filters, ICs, and other components from Burr
Brown, Crystal, et al, dCS has designed their very own patented D-A conversion architecture called a dCS Ring DAC®. I’ll
get into what specifics I know about the device later in the
review, but for now what’s important to know is that dCS developed the technology, which eventually turned into the Ring
DAC, as part of their work designing high performance A/D
converters for British military airborne radar systems. The
accuracy of these systems was the difference between a successful mission and big splat.
The reasons that high-end manufacturers don’t make their own
D-A converters are the expense and level of expertise involved.
A lot of R&D, and a lot of money. Consumer electronics giants
Sony and Yamaha, both of whom use their own LSIs for D-A
conversion in some products, are the only other companies I
can think of who make their own DACs. High-end companies
typically don’t have those kinds of resources, financial or otherwise. Due to their past in the defense industry, dCS does and
that is a significant distinction for their products.
Verdi Transport
The Verdi CD/SACD transport is based on a Sony broadcast
standard transport mechanism that costs a lot more than the
mechanisms used in most players. How much better that
makes it than a typical disc player I don’t know, but considering
that most consumer players use transport mechanisms that
cost just a few dollars (OEM parts cost) it’s obvious Verdi is a
different animal. This drawer loading mechanism is equipped
with dual lasers for CD and SACD. The drawer and mechanism
are suspension isolated for vibration control.
The Verdi has every brand of digital output known to man,
dCS Verdi & Delius
including three AES/EBU outputs, one each RCA, BNC, toslink
optical, and ST optical interface. There are two IEEE 1394 outputs for DSD data and wordclock inputs and outputs on BNC.
The 1394 designation defines the physical interface for the
DSD data stream—the security protocol is proprietary to dCS,
and the 1394 DSD digital output functions only with their gear
(no open high res digital standards yet!).
The wordclock ins and outs allow the Verdi, or a connected
DAC, to be used as the master clock depending on how the
connections are made (more on this in the setup section
below). There are also CH1 and CH2 data outputs (on BNC)
that can be used in conjunction with the wordclock output to
transmit 16/44 signals, which is more common in the pro world
(dCS has their toes firmly planted in both the pro and consumer sides of the biz). A 9-pin Software Upgrade Connector
for updates via PC is also included, and the power cord is
That last point is an important one. The Verdi‘s digital processing and control functions are software based and can be
upgraded with relative ease. Some upgrades can even be performed by simply putting a CD disc in the player to upload the
software changes. I think upgradeability is a critical feature for
a product at this level and price, and the Verdi clearly delivers
on that score. Verdi also ships with a 1-year “guarantee” that is
automatically extended to 5 years when purchased from a dCS
dealer and registered with dCS. What’s more, the guarantee is
fully transferable to subsequent owners who will then be
notified as updates are made available. How’s that sound?
Delius is in many ways a scaled down version of the $15K
Elgar Plus. Audiophile Systems describes the main dif ferences
between the two pieces as being the Elgar’s improved fit &
finish (although Delius obviously isn’t lacking in good looks),
and more advanced power supply and balanced output circuits.
Delius is also specified as a 24-bit/192kHz-DSD DAC, but
some explanation is required. Although its digital inputs accept
high res signals (up to 24-bit word lengths with samples rates
of 48k, 88.2k, 96k, 176.4k, and 192k) and it’ll certainly play
‘em, the Verdi transport (at least as my sample was configured)
recognizes only 16/44 CDs and SACDs. It would not recognize
24/48 or 24/96 tracks from the video zone of DVD-Video standard discs. Using a Purcell digital-to-digital upsampler you can
convert 16/44 to a “higher resolution” signal to send to Delius,
but you must use a different external transport (like a DVD
player) as a source for native high res PCM.
DSP-based processing that oversamples PCM signals to 5-bits
at 2.8MHz for 44.1k sources (and its multiples, 88.2k, 176.4k)
and 3.07MHz for 48k sources (and its multiples, 96k and 192k),
respectively, with noise shaping employed to shift noise out of
the audio band. Sounds familiar doesn’t it? Substitute the 5-bits
for 1-bit and it sounds a little bit (pun intended) like DSD, doesn’t it? On the surface of it, this isn’t much different than what
happens in other players. The players covered here by W adia
and Resolution Audio also use DSP-based oversampling filters,
only at different sampling rates than dCS uses. (See the
reviews of these other products in this issue.) The difference is
that, like the Elgar Plus, Delius uses dCS’ vaunted Ring DAC®
As opposed to the off-the-shelf DAC chips used in virtually all
other digital components, the dCS Ring DAC® is a whole
daughter-sized board full of discrete resistors controlled by
gate arrays. Ring DAC’s 5-bit architecture operates at 2.8MHz
and 3.07MHz and keeps the inherent linearity advantages of 1bit sigma/delta D-A systems while minimizing (or seemingly
eliminating) the disadvantages—namely clocking errors and
noise. Its resistors are exceptionally accurate and don’t drift
with time or temperature changes. Errors are randomized and
effectively cancelled by samples being taken from a broader
number of current sources. The results are (claimed to be) linearity and resolution that are virtually unknown in any other D-A
conversion system.
dCS Verdi & Delius
If you read Stereophile’s measurements of digital source components you know that the very best of them measure around
20-bits resolution at the DAC output, even if the DAC in question is specified as 24-bit. That’s why I generally specify a DAC
as being “compatible” with 24-bit signals instead of just calling
it a 24-bit DAC. Notice that I didn’t do that in this case. dCS
claims the Ring DAC® to be capable of output resolution of no
less than 24-bits (29.5-bits according to Audiophile Systems).
That’s a bold statement. Resolution this high is difficult to
measure—some existing test gear apparently wasn’t up to the
task so dCS designed and built their own!
So, a few questions still come to mind. The dCS Ring DAC®
operates at 5-bit/2.8MHz (or 5-bit/3.07MHz with 48k). What
does it do with DSD, which is 1-bit/2.8MHz native? Or with the
Purcell’s upconverted DSD stream? Is the signal decimated to
5-bits? I don’t know, but I can assure you that I asked dCS,
who wouldn’t tell me—trade secrets, you know. So, much as I’d
like to tell you what this player is doing to make such groundbreaking sound, dCS is apparently afraid to let the cat out of
the bag, lest their competitors take advantage.
Now that I’ve devoted so much space to the logical side of
things, let’s get the physical world. There is one set each single-ended (on RCA) and balanced outputs, Delius has two digital inputs each on AES/EBU and coax/RCA, one coax on BNC,
and one each toslink and ST optical. There are two IEEE 1394
digital inputs, and wordclock input/output on BNC. Also included is a digital output on BNC that will output digital signals up
to 24/96 resolution. Delius is also software upgradeable, and a
9-pin Software Upgrade Connector is included and like the
Verdi, the power cord is removable. Hardware changes are
sometimes required, but dCS makes them as inexpensive as
they can for their customers. Delius comes with the same
transferable warranty outlined above for V erdi.
Setup, Controls and Functions
The Verdi/Delius combo is made to stack. V erdi’s chassis is literally two Delius chassis stacked together, creating the overall
impression of three boxes on top of one another with the disc
tray in the top box. Aesthetically, the dCS stack is as good
looking as gear gets, in a slightly sci-fi way. A conversation
piece(s) nevertheless.
Not only does the Verdi/Delius present a dazzling array of fea-
tures, dCS also offers exceptionally well-written and understandable user manuals and even goes the extra mile by of fering quick setup guides that show how to use each product
together with dCS’s other products. The dCS Verdi with dCS
Delius guide shows how to make the recommended physical
connections, and how to engage the logical factory settings.
For simplicity’s sake I followed the quick setup exactly, connecting the AES 1 and 1394 digital outputs of the Verdi to the
AES 1 and 1394 inputs on the Delius, along with connecting
the wordclock “out” of Verdi to Delius’ wordclock “in.” With the
exception of the wordclock connection, this worked out famously.
Verdi and Delius operate at two clock speeds. One clock is
used for 44k sources and its multiples, 88.2k, 176.4k, and DSD
at 2.8MHz. The other clock speed operates for 48k and its multiples, 96k and 192k. As mentioned, either Verdi or Delius can
be used as the single, master clock for the entire system. The
quick setup guide recommends a connection method that uses
Verdi’s clock as the master, which is how I used the system for
a few weeks before I began putting this review together. A closer look at the Delius user manual and the dCS web site
revealed their recommendation that Delius be configured as the
master clock to reduce jitter (and increase sound quality). I
tried this configuration and did hear perhaps a bit better sound.
In any case, maybe amends should be made on the quick
setup guide.
The dCS stack then functioned as the ultimate plug & play—
putting a 16/44 CD into Verdi automatically switches Delius to
the active AES 1 input, while putting an SACD in switches to
the 1394 input and vice verse. Unlike a trip to the dentist it’s
quick and painless! Hybrid SACDs automatically boot to the
(two-channel) SACD layer, unless instructed otherwise in
Verdi’s “favorite layer” menu selection. Indicators on Verdi let
you know what kind of disc is playing (CD, SACD, hybrid
SACD, etc.) and show the title and track text programmed onto
most SACDs. The display is difficult to read at a distance.
The front panel display of Delius tells which input is active and
by default displays the bit depth and sample rate of incoming
PCM signals and merely shows “Verdi” with DSD. The only
aspect of interacting with the dCS stack that isn’t a snap in
some respects is the remote control. Although it’s one of those
substantial and heavy-duty aluminum jobs, direct track access
is a little weird. You have to hit one button to activate the number keys, and then you need to enter the number zero before
the track number, even before a double digit number. To make
dCS Verdi & Delius
a single digit track access quicker, two zeroes are required.
Once a disc is playing, the choices increase further. For CD
and SACD there are four separate filter algorithms that can be
selected, and for higher resolution PCM two more filter settings
become available. Delius remembers the last filter setting used
for each format, and for each PCM sample rate specifically.
The filter settings can be implemented from remote control as
well as the front panel display menus. For PCM I chose Filter
4, which offers the gentlest low-pass slope with the greatest
time domain accuracy. For SACD, after some experimentation I
ended up with Filter 2. The user manual offers no specifics on
its characteristics beyond noting that Filter 1 offers the widest
bandwidth and Filter 4 is for measuring, not listening. The concern with SACD is that the high frequency noise might damage
wide bandwidth preamps, amps and/or speakers. Maybe it was
just a sense of cautiousness playing with my mind, but I
thought Filter 1 sounded a bit hashy in my system.
Trust me that what’s described above is not even a summarization of Delius’ functionality, but a mere mention of just some of
its capabilities. Among other things Delius can also drive
amplifiers directly from its analog outputs, and comes with a
fully functional volume control. I didn’t have a chance to fully
evaluate this aspect of its capabilities as I’d been able to spend
only a few weeks with it at press time. I plan on writing more
about this product, and Audiophile Systems has mentioned that
I might get an opportunity to use a Purcell with the V erdi/Delius
stack. Either way, look forward to reading more about dCS’
gear in future issues of the Journal.
From the first moment I loaded an SACD into the dCS stack I
knew I was hearing more layers of resolution and clarity from
my favorite SACDs than I’d ever heard before. It was stunning.
Front to back, side to side, I felt like I was being taken farther
and farther into the recording space. I’ve never heard any
source component with more soundstage size or focus beyond
the speaker positions. Tonally the dCS stack is extended at
both frequency extremes and resolved in the midrange, but
totally neutral as far as I can tell. What goes in comes out, with
no unwanted emphasis anywhere. As thrilled as I’ve been by
SACD sound in general, I didn’t know just how much more
information was there on the discs, waiting to be retrieved.
Virtually everything I admire about the SACD playback I’ve
heard from other players was there—the rhythmic “rightness,”
the natural detail and ease, the three-dimensional and convincing images—but I heard more of everything with the dCS stack.
Before reviewing the dCS stack I’d regarded SACD as a close
runner-up to good vinyl playback, and had yet to be convinced
that DSD could sound better than the very best analog. SACD
doesn’t sound digital in any way on the better players (particularly the Sonys) and shows an obvious and vast improvement
in resolution of low level musical details and spatial accuracy
and dimension (and anyone who tells you this isn’t the case is
really telling you a lot more about their system than the SACD
I’ve regarded SACD as good enough that the convenience factors of an optical disc system started to make sense, but I still
felt I was hearing more air, transparency and overall involving
sound from vinyl. There has to be something going on if I’m still
willing to go through the ritualistic abuse required for playing
records! My experiences with the dCS gear and a new recording by the Eighty-Eight’s label have changed my perception of
SACD’s place in the recording media food chain for the first
Clark Terry and Max Roach’s Friendship was recorded simultaneously in DSD and analog. I compared the SACD version
played on my Sony DVP-9000ES SACD player and the dCS
stack against the vinyl playback version on a Linn Sondek
LP12/Linto combo. Comparing both versions of this recording
showed me two interesting things. First, I couldn’t hear
significant differences between the LP and the SACD when I
played the SACD on the Sony, which in itself is impressive.
Second, playing the SACD on the dCS stack, I heard some
significant differences between it and the LP, and the SACD
sounded better!
Friendship is an audiophile’s dream—sensational music that’s
recorded to the highest aesthetic standard, with sound that’s
just amazing. Although there are a lot of great tracks here,
“The Nearness of You” (track #13) is just stunning in its ability
to conjure the image of Clark Terry playing his horn in my
room. The nearness of him on this recording is unbelievable,
an almost physical presence. The image is rounded and holographic with more convincing air than CDs ever show. You can
hear Terry inhaling before every breath into the horn and the
keys tapping into the instrument’s body. Other quirky room
dCS Verdi & Delius
sounds are apparent, enhancing the illusion of being in the
recording space. The air into and out of Terry’s horn has the
most natural dynamics I can imagine, and that’s one of the
ways in which the DSD recording steps above the LP.
The excellent sounding LP is softer, slightly muted and somewhat compressed in dynamic contrast compared to the SACD.
Terry’s mini-crescendos keep going and going, and on the
SACD those peaks hit a little higher in level with more openness and impact. For the first time, I had the distinct sensation
that DSD could deliver playback that’s more transparent than
analog and freer of any form of coloration. That the SACD
could better such a fabulous sounding LP was remarkable, and
it took this dCS playback system to reveal that superiority.
Listening to SACDs through the dCS stack is consistently revelatory. Ray Brown’s acoustic bass on Groove Note’s SACD of
Soular Energy is fat and natural, with rollicking pace—SACD
gets the timing of the music just right. With Beck’s Sea
Change, the degree of texture and shading of the instruments
and Beck’s vocals are mesmerizing. Yet another standout is
Analogue Productions’ SACD of Bill Evans Trio’s Waltz for
Debby. Although the mix is very stereo—as in very left and
right with little in between—this SACD has a startling live quality that captures the club sounds, as well as an incredible
amount of airy detail in the cymbals and snare brushes.
Whatever SACDs I played, the dCS stack communicated to me
more of the music and the feelings behind the music.
With CDs the dCS stack is no less impressive, delivering massive but precisely focused soundstage, and layering the musicians on distinct planes front to back. As with SACD, the program material and the other gear in your system will determine
the boundaries of the soundstage. The degree of clarity is striking, and the sound is highly resolved with excellent extension
at the frequency extremes. It’s very transparent to the source
and yet there’s a good degree of musicality as well, especially
with the Filter 4 setting. The resolving power of the dCS stack
is right up there with the Ayre D-1x and Wadia 861 as the most
revealing CD playback rigs I’ve heard.
In my opinion, even with SACDs, the dCS stack leans just the
teensiest hair to the analytical side. Buried deep in all its transparency is the slightest hint of coldness, a condition that is
more noticeable with CD playback. As with A yre’s D-1x, I wonder if this system is simply revealing more about the source
material than I care to hear. (I also wonder how far a Purcell
upsampler might go toward eliminating this perception.)
Certainly that’s a possibility because my perception did change
with 24/96 material from an external source. Also, the dCS
stack doesn’t create the palpably holographic and convincing
images delivered by the Wadia 861. The Wadia’s images are
simply more cohesive, rounded and convincing—more like analog in that respect than any CD player I’ve ever heard. Note too
that I’m talking about the slimmest of margins here. The W adia
and this dCS stack are the two most astonishing playback systems I’ve heard with respect to combining high resolution and
convincing musicality and imaging. I’ve not heard a playback
system with more resolution than the Verdi/Delius combo, and
only the Wadia eclipses it in musicality. This is one of the finest
CD playback systems available, period.
The dCS Verdi/Delius is a state-of-the-art playback system and
that is not the only important thing going on here. This dCS
system didn’t just reveal itself as an outstanding playback system. It revealed more of the vast potential of the SACD format
itself than anything I’ve yet heard. The Verdi/Delius rig represents the best SACD playback I’ve heard and is competitive
with the very best CD playback systems available at any price,
even if it falls slightly short of the clear-cut benchmark status it
has established with SACD.
Judging by what I’ve heard at trade shows, the addition of a
Purcell to the Verdi/Delius system might change this perception
and I hope to report to you on that in the future. And this isn’t
even dCS’ flagship system. The Delius sounds so good that it’s
difficult to imagine what the Elgar Plus sounds like! Once a certain level is attained incremental improvements in audio quality
often come with disproportionate price tags. Considering that
it’s still somewhat commonplace for CD playback systems to
cost upwards of $20K without the ability to play SACD and/or
any form of upgradeability, I think the dCS offers performance
and flexibility commensurate with its admittedly gaudy price.
Even at $21,995, the Verdi/Delius stack is what I would have if
I could afford it. APJ
dCS Ltd.
US Distributor: Audiophile Systems Ltd.
8709 Castle Park Dr. Indianapolis, IN 46256
www.audiophilesystems.com 888 272 2658
ne player in this survey stands apart from the others. The
Musical Fidelity Tri-Vista CD/SACD player is unique in a
number of ways. It will play CDs, SACDs, and high res music
on DVD-Video standard discs (DADs)—it’s the only player that
we’ve seen without a video output that will play DADs. It has a
choke-regulated power supply with multiple transformers and
uses miniature vacuum tubes in the output stage. It provides
near state-of-the-art performance for both CD and SACD playback, and it’s reasonably priced for a product with this build
quality and level of sophistication.
The Company
Musical Fidelity is a UK company founded in 1982 by a classically trained clarinetist named Antony Michaelson. MF makes a
variety of high-end audio components and has established an
excellent reputation for performance and value. Michaelson
demands that all MF products deliver musically satisfying
sound and the samples that I’ve heard do.
The Tri-Vista CD/SACD player is a limited edition component
with analog circuitry that utilizes miniature vacuum tubes. Only
800 Tri-Vista players will be made.
Hi-fi components made in the UK are often more compact than
their Japanese or American counterparts with industrial design
that could be flatteringly described as subtle. The T ri-Vista is
neither compact nor subtle. It’s a brawny beast with bold
styling. A very attractive component in my opinion.
The unit is quite large, weighing more than 50 pounds, and is
beautifully finished with a thick silver face plate and black fluted
sides. The front panel has six buttons to control transport functions, plus a power switch. The disc tray is centered just below
the bright display, which shows the usual information and can
be easily seen from across the room. The chassis sits on four
large feet with translucent rings in the middle of each. These
rings glow red while the power supply stabilizes immediately
after the unit is first turned on, then fade to amber as the unit
un-mutes and becomes operational. After about a half-hour,
when the Tri-Vista has completely warmed up, the glowing
rings around the feet turn blue making the product seem to levitate above the shelf. This trick looks unbelievably cool and
everybody who witnessed it at my house was suitably
The Tri-Vista rear panel is very simple. There are coaxial and
optical (toslink) digital inputs and outputs (one each). A single
pair of single-ended analog outputs is provided on RCA connectors and the power receptacle is an IEC-type allowing
power cord replacement. The remote control allows the user to
access all player functions.
The Tri-Vista
uses a Philips
with dual
DACs have
differential output but MF, like many UK manufacturers,
eschews balanced circuitry and provides only single-ended
analog outputs. Two Crystal 8420 sample rate conversion
(upsampler) chips are incorporated in what I presume is a cascaded array of filters. MF claims that the signal paths for CD
and SACD are separate but offers no additional information.
The Burr-Brown 1738s are claimed to be compatible with PCM
and DSD data so perhaps one is used for each signal type.
The power supply in the Tri-Vista is very impressive indeed.
Multiple transformers and choke filtering are employed. The
power supply in this disc player looks more substantial than
what you’ll find in many amplifiers.
The output stage incorporates four miniature 5703 mil spec
vacuum tubes. A small amount of negative feedback is utilized
in the analog circuitry.
Wadia 861
Sound and Performance
I listened to two samples of the Musical Fidelity T ri-Vista
CD/SACD player, both borrowed from owners because there
wasn’t time to obtain a review sample from the manufacturer.
Both samples performed in a consistent manner and delivered
excellent, satisfying sound. The Tri-Vista offers uncompromised
CD and SACD performance that rivals the best players we’ve
heard for each format.
With regular compact discs the Tri-Vista presented sound that
was highly detailed yet smooth. Images were well defined and
three-dimensional with excellent focus. The sound was vivid
with just a touch of character in the upper midrange.
For the first several days I perceived an upper midrange push
bordering on aggression. After a few days of continuous play
this settled into a very pleasant musical presentation with just a
touch of a distinctive thumbprint added by the player. The presentation was always exciting and musical with what I would
describe as a slightly “technicolored” enhancement.
The Wadia 861 is slightly more transparent and neutral and
seems to retrieve slightly more detail from CDs. The Wadia creates a more palpable image with even more depth but the T riVista delivered sound that was emotionally engaging, very
dynamic and nearly as refined. I preferred the Wadia overall
but the Tri-Vista was not far behind and it also plays SACDs
and costs about two grand less.
SACDs were presented with a similarly enhanced sound.
SACDs sound a lot better than CDs and that fact was clearly
evident when comparing the formats on the T ri-Vista player.
SACDs allow the top end to open up providing a more natural
presentation of air and space and the MF player got all the
musical information right. The sound was fuller and more substantial than what the Sony 9000 can provide and slightly more
refined, although the Sony, which costs far less, was not
embarrassed in the comparison. The dCS stack, costing more
than 3 times as much as the MF player, delivered sound from
SACDs that was slightly more transparent but the Musical
Fidelity sound, while not as neutral and colorless, was perhaps
slightly more musically engaging.
performer. It can stand up to the very best CD players in headto-head competition and it rivals the best SACD player I’ve
heard. It has a slight character of its own but is never of fensive
while providing a listening experience that is always fun and
exciting. At $6000 it’s a bargain and a benchmark that sets new
standards for performance and value. APJ
Musical Fidelity
USA Importer: Kevro International
902 Mckay Road #4, Pickering On, L1W 3X8 Canada
001 905 428 2800
ver the last three years I’ve spent a lot of time listening to
disc players at my home and have brought many of them
to Richard Hardesty’s listening room to listen to them in his
system and compare them with one another and with Dick’s
reference gear. During that time, Dick’s reference CD player
has been Wadia’s 860, which was then upgraded to the 861
(the 861 sells for $7,950). During this time many players have
come and gone; few have even approached the Wadia’s performance, and none have unequivocally surpassed it. In fact,
I’ve yet to hear a single player that I prefer overall to the Wadia.
Let’s take a look at this reference quality player.
Design and Construction
The heart of Wadia’s digital technology is the DigiMaster™ filter
system. This software-based filtering solution is implemented
on high-powered dual Motorola DSPs. Software-based solutions are inherently flexible, allowing relatively easy running
changes and improvements. The 861’s DigiMaster version is
2.4, which engages 32x oversampling with 44.1k and 48k signals.
The 861 is also compatible with signals up to 24/96 resolution
(oversampling at 16x with 88.2k and 96k sources), but, since
its drive is a CD mechanism, signals with resolution higher than
The Musical Fidelity Tri-Vista CD/SACD player is a remarkable
Wadia 861
16/44 must come from a player with a DVD transport via the
861’s digital input. Wadia makes a point of oversampling at a
direct multiple of the incoming sample rate as they believe that
asynchronous oversampling (oversampling a 44.1k signal to
96k instead of 88.2k, for example) has deleterious effects on
sound quality. I don’t have such a rock solid opinion on this philosophy. I’ve heard some products that perform asynchronous
oversampling that sound bad, but I’ve also heard others that
sound excellent, such as Musical Fidelity’s T ri-Vista.
The 861 uses two K-Grade Burr Brown PCM-1704s per channel in a dual differential stack for each channel. (The 860 used
Burr Brown PCM-1702s, and I’ll explain the differences I heard
between the two iterations later in the review.) Following the
DACs the 861 uses Wadia’s proprietary Swift Current technology. Based on a “current conveyer” IC (Integrated Circuit), this
solution performs current-to-voltage conversion without negative feedback.
Many DACs incorporate their own I/V (current-to-voltage) conversion with op amps that employ negative feedback. The players I’ve heard that avoid feedback in the analog output section
image in a more coherent and convincing fashion than nearly
all the players I’ve heard that use feedback. Whether this is
evidence of feedback introducing timing errors (feedback is a
time domain phenomenon) or simply that the I/V conversion
internal to DACs is generally poor I can’t say for sure. That the
Wadia eschews feedback and sounds this good probably
means something though.
The 861 has user-selectable low-pass filter algorithms with dif-
ferent roll-off characteristics. The setting that Dick and I both
prefer is algorithm “A,” which has the gentlest roll-off and the
best time domain performance, resulting in the most organic
and musically satisfying sound.
The 861’s analog output stages are truly balanced, maintaining
balanced operation from the DACs to the output. It also has a
“The Wadia’s construction
materials are heavy-duty, and
I mean that literally.”
sophisticated digital volume control and high current output
stage that allow the 861 to drive an amplifier directly as a digital preamp.
The Wadia’s construction materials are heavy-duty, and I mean
that literally. The 861 tips the scales at 48 lbs. and its remote
control is heavier than a lot of players on the market. It has a fit
& finish and overall look and feel that instill the kind of pride of
ownership one expects from an $8K CD player. The body is a
“monolithic” machined aluminum piece and the disc drive
mechanism is a trick job by Teac with its own internal clamping
mechanism to minimize vibration and a self-aligning digital
servo to adjust tracking errors. Wadia claims that this “full”
clamping mechanism offers superior performance even to toploading mechanisms that use a manual clamp that fits only
around the disc’s spindle hole.
The 861 uses two separate power supplies with their own
toroidal transformers. One supply is dedicated to the transport
mechanism and servo, while the other powers the D-A circuits.
The transformers are shielded in an aluminum subchassis with
neoprene lining on the top and bottom of the enclosure to prevent any hum, noise, vibration and other transformer-related
artifacts from contaminating the audio circuitry. As Dick’s experience of upgrading his Wadia 860 to an 861 indicates,
upgradeability is a key tenet of W adia’s philosophy.
In addition to upgradeable software-based DSPs, the W adia
861 is physically constructed in a modular fashion with separate, replaceable boards for inputs/outputs, DACs, display/controls, and the transport mechanism. The rear panel is also easi-
Wadia 861
ly replaceable to facilitate new interface options. (The same is
also true of Wadia’s more affordable 301 CD player.) Wadia
takes pride in making sure your cheese isn’t left out in the
changing winds of the digital world!
The Wadia 860 was smooth, refined, and musically involving in
a way that nearly all CD players just aren’t. Images were full
and round with superb spatial delineation. But, compared to the
most detailed and resolving players (like A yre’s D-1x), the 860
was noticeably soft and warm. And I don’t mean to imply that
was a bad thing. Playback of 16/44 redbook CDs can benefit
from the right amount of softening around the edges. Over the
long haul I’ve often felt it would be more pleasurable not to
hear as many of the CD’s flaws (particularly the hard, glassy
sound that bright recordings exhibit) even if that means losing
some degree of resolution. The Wadia 860 was still competitive
with the best players out there, and while I may have found it
preferable in some ways to players with more resolving power,
it was obvious that there were sonic tradeoffs for the “listenability” and musicality I enjoyed so much.
Being so familiar with other products that use Burr Brown’s
PCM-1704 DACs, Dick and I both expected that the 861
upgrade (which switched out PCM-1702 DACs for the PCM1704s, among other things) might be the cat’s pajamas if it
bridged the gap between the 860’s musical righteousness and
the more highly detailed response of players like the A yre.
Man, did it ever work out that way!
The revamped 861 has it all—extended response at the frequency extremes, a midrange with incredible resolution of low
level musical detail and texture, and a grand soundstage with
precisely drawn layers of image focus. The 861 retained all that
was engaging about the 860 and added more of everything.
The 860 got the music right, but the 861 can take you into the
recording space in a way its predecessor couldn’t. And it does
so with a deliberate grace, making instruments and vocals feel
more natural, real and vital than any other CD player I’ve
heard. Although dCS’ Verdi/Delius stack and Ayre’s D-1x at
least match the 861’s resolving power, they don’t quite match
its vitality. Images of musicians and vocalists are fuller, more
three-dimensional and more holographic. It’s as though there’s
more air around the images, which have mesmerizing solidity
and substance.
Tonally, the 861 is rich, full and warm but doesn’t lack any
degree of air or extension on top. It captures the full foundation
of the music, along with all the midrange and top end detail,
without the slight leanness that accompanies the Ayre D-1x’s
resolving capabilities. The 861 also has a remarkably deep
soundstage, front to back, as well as incredible ability to focus
far out to the sides of the stage. At all times even the subtlest
differing planes are revealed to create most convincing images
of musicians in front of you, spatially differentiated from one
another and clearly defined.
All of the 861’s considerable attributes improve commensurately when a 24/96 signal is fed into its digital input. 24/96 is much
more rhythmically right than redbook CD, with greater resolution of the timbres of instruments and the recording space. The
ability to focus instruments way out to the sides and far back
behind the speakers is particularly improved with 24/96 and the
Wadia does more justice to these recordings than any player
I’ve heard.
The Wadia 861 is one of the very best CD players available at
any price, and is unquestionably the one that I enjoy listening
to the most of all the players I’ve heard. The 861 is one of the
players that you should look at very closely if you have the
$8000 to spend on a CD-only player. Musical Fidelity’s $6K T riVista, for example, is nearly as good a CD player and not only
plays SACDs, but is a cut above the Sony players with SACD
performance. That being said, for those who demand the best
and can afford it, the Wadia 861 is an absolute killer CD player.
Its pure performance edge and W adia’s commitment to
upgrades for its customers make this player a compelling
Special Note: As we were going to press Wadia informed us
that a special edition version of the 861 is about to become
available that offers an even more advanced transport mechanism and makes other incremental improvements to the player.
We can’t wait to hear it, and will be reporting on the 861 SE in
a future issue of the Journal. APJ
795 Highland Drive
Ann Arbor, MI 48108
www.wadia.com 734 975 4217
ere are just a few words on how these reviews were conducted and the software frequently used to evaluate the
players. Much of the review work was done by me at my house
in northern California, and I also brought many of the players
with me to Dick’s home in southern California so that we could
listen together in his system.
My system includes a BAT VK-51SE tube preamp, and
Audioquest interconnects and speaker cables. For the past
several months, I’ve been using Arcam’s FMJ DV27 as a reference for CD playback and have also used Sony’s DVP-9000ES
and XA-777ES SCAD players. Dick and I both use V andersteen
3A Signature loudspeakers and 2Wq subwoofers high-passed
with Theta Dreadnaught power amplifiers. We both use two
multichannel Dreadnaughts in a vertically biamped configuration for each front channel. We both use Linn Sondek LP12
turntables, although Dick’s is more up-to-date than mine. While
his is a full Lingo/Arkiv/Ekos rig, mine is a 1985 vintage table
with the fluted body and original power supply, Ittok tonearm
and stock tonearm cable. I have updated to the Arkiv cartridge,
and I use Linn’s Linto phono stage. Dick uses the phono stage
in his Ayre K-1x preamp. Both our rooms share double duty as
home theater environments, and Dick’s is acoustically treated
more extensively than mine. (I couldn’t get away with any more
while maintaining a harmonious household.) I’m about to move
into a new house that has a completely treated, dedicated listening room/theater (can’t wait!).
I’ve been to Dick’s place so often and listened to his system so
much that it really is a second reference system for me. My
opinions here are amalgams of my experiences in my room,
gathered over several weeks or months of listening, combined
with relatively infrequent but intense sessions listening with
Dick in his room. For example, during a three-day period we
listened to the Ayre CX-7, Resolution Audio Opus 21, and the
dCS Verdi/Delius stack. We compared these players to Dick’s
Sony DVP-9000ES and Wadia 861 players, and to one another.
We generally compared two or three players at a time, switching between inputs on the Ayre preamp. We typically listened
for longer periods with a lot of material to gather our initial
impressions before focusing down to switching between players
with one or two cuts to discern and quantify the subtlest dif ferences between the players. We took frequent breaks from listening to avoid fatigue.
In my own listening room, I typically listen to a component over
days, weeks, or even months. During that time I respond to it
emotionally or I don’t. In either case, after that’s established, I’ll
do some switching between the review component and a reference (or references) to quantify the differences between them
with distinction. Neither of us performs blind testing because
we both feel that lends itself to relying too much on the analytical side of the brain, which is the opposite of the side of the
brain that music appeals to. As simple as it seems, listening to
the music and seeing if you tap your feet or get goosebumps
on your arm can tell you a lot.
System Revealing Software
I thought many of you would like to know what we listen to, and
what we listen for when evaluating equipment. As I started putting this list together and speaking with Dick about it, this
became a more puzzling effort than it initially seemed. Dick and
I both respond to image focus, specificity, and palpability. So
many of the best recordings for revealing the performance
attributes of an audio system are small ensemble pieces and
jazz recordings in particular. Although I like jazz very much, it’s
not the only thing I listen to, and in fact doesn’t even embody
the majority of my listening. And, when we’re reviewing components, Dick and I both do what we always do: we listen to
music we like for pleasure. All kinds of music.
We listen to classical and jazz, classic rock and new rock,
some folk and bluegrass, and virtually all things in between. My
quandary for this article was whether to write about music I
like, or music that I can recommend to our readers that will
allow them to determine if they can hear all the things in these
recordings we’re hearing. The answer is the list below that represents some of both.
The fact is that like most of you I listen to a lot of music that
isn’t “audiophile quality” in any way. And I’m still just as moved
by it. Listening to these recordings on a revealing system might
not be something I can wax poetic about in the fashion that’s
sometimes useful in communicating how a product under
review sounds, but I can assure you that some of my poorer
recordings connect me to the better components I review.
One of the recordings I own that will never make a list of CDs
with great sound is The Boatman’s Call by Nick Cave and the
Bad Seeds, and I’ll never forget the first time I heard that CD
on a revealing digital audio rig. It still wasn’t Belafonte at
Carnegie Hall, but so much more of the feeling of the words
was communicated to me through the inflections of the vocals.
Cave writes the songs, sings, and plays piano, and his lyrics
and playing took on an even more profound resonance. I’m
getting the same goosebumps on my arms and neck that I had
that evening as I write this.
Neil Young’s recordings affect me the same way. His music and
lyrical poetry mean a lot to me; they’re evocative to me in
almost the same way that well written short stories are. While
they contain none of the story arc that novels and movies can
portray, they present clear pictures of emotion, feeling, relationships, and situations. Among the many things I admire in his
work is his ability to re-interpret his decades-old tunes and
make them fresh again, and often simpler.
Neil Young Unplugged features the best takes I’ve heard on
“Pocahontas,” “The Needle and Damage Done,” and
“Helpless.” This mostly poor recording was done on the spot at
Universal Studios for MTV. I know that it means a lot to me
when a component can give me just a little more of the guitar’s
sound, the body and hands on the strings, and more air in
Young’s harmonica. But I’m not sure there’s enough to say
about that to give you anything meaningful to listen to in order
to evaluate your system.
And then there’s the issue of genre. Fanatics of both classical
and jazz can be Nazis about whether reviewers listen to their
brand of music to evaluate gear. Classical by itself can certainly reveal a lot about a system’s dynamic capabilities, its tonality, and its ability to transport the listener into a very large
space. But people who listen exclusively for which gear best
recreate the air and space of Carnegie Hall can often end up
choosing components that don’t provide the kind of focus that
better serves other, more intimate forms of recorded music.
Jazz suits me not only because I like it, but because the often
intimate nature of its recordings is revealing in a way that lends
itself to a number of different genres.
I guess what I’m getting at here is that we listen to music that’s
all over the map; we’re just eclectic kind of guys. Many, many
forms of music help us to form our opinions, but not all of them
lend themselves to being written about as reference material.
Trust though that no matter what your bag of musical taste is,
we listen to it and it does enter into our opinions even if you
don’t find it mentioned in our reviews or lists of reference software.
This initial list includes a few digital recordings you might want
to try in your own systems and listen for some of the attributes
described here. We will be updating and adding to our software
lists from time to time, both in print and on the web site. I’m
starting here with just some of the recordings Dick and I both
use of music we both like. Note that I’m sticking with digital
because that’s what was reviewed in this issue. Get ready for a
truckload of vinyl recommendations in the next Journal!
Mark Levinson Live Recordings at Red Rose vol. 1–SACD
According to the liner
notes, this pure DSD
recording was made
by Mark Levinson (the
man, not the company) in his Red Rose
Music store in New
York City using two
microphones, a preamp and a Sony DSD
recorder. No editing,
no jive of any kind. The result is perhaps the most stunningly
natural set of recordings I’ve ever heard. The entire disc is full
of tunes that I enjoy listening to from start to finish, and still do
even though this disc is my most often-used demo piece! Here
are some of the highlights.
It opens with Chico Freeman on tenor sax doing a riveting,
breathy interpretation of Duke Ellington’s “In a Sentimental
Mood” accompanied by George Cables on piano. You can hear
every part of the sax, from Freeman’s mouth blowing through
the reed (sometimes in a spitty fashion) to his fingers working
the keys. And you can clearly hear (if not practically see) him
shifting on his stool while he plays.
There are three solo piano pieces that are riveting: “Misty,” “It
Might as Well be Spring,” and “Alligator Crawl.” The piano
sounds unbelievably lifelike, especially the weight of the lower
registers. You can hear the player’s fingers on the keys, and
there’s amazing flow to the timing of the music and the pedal
work. The piano decay seems to just float onward and upward
forever. On high resolution systems you’ll not only hear the
players humming along with the tunes or clearing their respec-
tive sinuses, you can hear some traffic noise and other ambient
clues that take you beyond the music and into the Red Rose
There’s also a killer reading of a Rupert Brooke poem, “Little
Dog’s Day,” by Kim Cattrall accompanied by Levinson himself
playing an old Italian bass. A very distinct image of crickets outside the house is drawn way up toward the ceiling, about
halfway between the left speaker and the listener—you’ll think
they’re outside your listening room! Levinson’s double bass has
weight, and you hear the cavity of the instrument as well as the
strings and its snappy pace. Cattrall’s voice is warm, full of
subtle inflection and very natural. This disc dramatically
demonstrates the superiority of DSD over CD and is an
absolute must have for any audiophile—it’s also a hybrid disc
with a CD layer, so it’s mandatory even for those without an
SACD player.
Ray Brown Trio Some of My Best Friends are the Trumpet
Players–Telarc CD
Ray Brown was a world-renowned jazz bassist who played with
many of the greats, and always put together first-class musicians to play on his records. This set features trumpet players
like Terence Blanchard, Nicolas Payton, and Clark Terry,
among others. The whole disc rocks but I listen most often to
track 3, which reveals the full measure of Brown’s nimble work
on the acoustic bass. You can hear the body and strings, and
Brown’s hands sliding up and down as he plays. You can hear
how well your woofers are integrated with your main speakers,
with the right amounts of weight and speed when your system
is dialed. The piano is set off to the left, on a plane slightly
behind the bass, and the drum kit stretches out to the right
from the middle of the stage. The instrument images and timbres are natural, and the horns just rip out of the background at
a slightly higher point in space. On the better players/systems
you can hear the trumpet player blowing air into the horn and
other low level details. This is a terrific recording that’s sourced
from a direct-to-DSD recording down-converted to 16/44 PCM
by Sony’s Super Bit Mapping Direct process. Unfortunately,
Telarc has told me that they don’t plan to release this recording
on SACD because it’s not a multichannel master. I hope they
come to their senses! I’d love to hear this disc on SACD, but
until I do I’ll just have to recommend this one on CD.
Lyle Lovett Joshua Judges Ruth–MCA Records CD
The jacket of this CD indicates this is a full digital recording.
The music has emotional resonance and
the sound is wonderful. Pianos and strings
have natural foundation and decay. Russ
Kunkel’s percussion
distinctively punctuates the music, with
cymbal splashes moving throughout the
soundstage. In addition to the moving (and often morose)
lyrics, Lovett’s voice and the way it’s recorded here can tell you
a lot. You can hear not only his voice, but the microphone and
his proximity to it (especially on track 8). You can hear his
intake of breath before he sings a verse, and even the parting
of his lips. On brighter players his voice loses some depth, and
the details noted above take on a spitty quality. Great music,
great sound.
Clifford Jordan Quartet Live at Ethell’s–Mapleshade CD
Mapleshade’s recording of Clifford
Jordan’s band is
charged with the very
essence of the club
this music was
recorded in. You can
hear some buzz in the
onstage electronics,
tinkling glasses and
murmurs in the crowd.
The musicians are very clearly drawn in stage, on noticeably
differing spatial planes front to back. The piano and cymbals
have life and air, and the recording of Jordan’s tenor sax is
stellar. Like Levinson’s recording of Chico Freeman, you hear
every part of the performance, from Jordan’s air through the
mouthpiece, to the body and keys. Remarkable resolution and
feeling, one of the best CDs I’ve ever heard and the performance of Jordan’s band matches it with soul-felt intensity.
Beck Sea Change–Geffen/Universal SACD
Beck’s Sea Change has hardly left my SACD playing rig(s)
since I bought it a little over a month ago. This disc sounds as
good as a fairly production-heavy pop recording can sound,
and I think artistically this is one of Beck’s most compelling
Paula T. Hardesty, Publisher
Richard L. Hardesty, Editor
Edith Hardesty, Copy Editor
Shane Buettner, Equipment Review Editor
Rick Johnson, Art Director
efforts—right up there with Odelay, my favorite of Beck’s previous releases. Beck’s lazy vocals and the guitars sound warm
and natural with sensational liveliness. As densely layered as
the mix is, this SACD never loses focus of any of the instruments. The drums and percussion in particular have a rich,
fleshy texture. On track 6 there’s a shaker that’s delivered with
mesmerizing clarity—you can practically see (and feel) the little
beans moving inside their shell with the music. This is one of
the rare pop recordings that manages to put the musicians in
your room—great music bordering on audiophilia.
above, this recording captures Krauss’s strings and vocal work
in particular, which is the heart of the record as one would
expect. There’s an incredibly natural feel to the guitars, with the
body cavities vividly portrayed and the lid completely lifted with
the strings’ decay. It’s very open and involving, and the emotional weight and quality of the music is phenomenal. Krauss’s
vocals are appropriately warm and cozy, which fits her style.
The highlights for me are “Maybe” and “Ghost in this House.”
It’s a gift to have a recording this inviting of such enjoyable
Alison Krauss Forget About It–Rounder SACD
Alison Krauss, a
frontline foot soldier in the latest
bluegrass movement, lands on
the pop side of
country folk with
Forget About It
and absolutely
pulls it off. As
with the Beck
release noted
Nojima Plays Liszt–Reference Recordings CD
Dick introduced me to this recording by calling Franz Liszt the
Eddie Van Halen of his time. This music, composed by a blazingly talented piano virtuoso, is certainly intended to be played
by one. And Minoru Nojima is the one! This is an excellent natural sounding recording of a piano. It captures in great detail
the percussive power of the instrument along with the authority
and speed of the piece and Nojima’s ability to express it to us.
The piano covers the broadest area of the frequency spectrum
of any instrument and, with this particular recording, you’ll learn
a lot about your system’s abilities at all ranges. You’ll hear low
level details—the sound made as the keys are struck—and the
wild variations in intensity of the notes that are played. APJ
Audio Perfectionist Journal TM (ISSN1525-3392) is published quarterly by Auricle Publishing Inc., 17141 Los Robles, Fountain Valley, CA. 92708-2027,
1-714-968-9405, email: [email protected] Bulk Postage Paid at Atlanta, GA. and At Additional Mailing Offices. Postmaster: Send address
changes to: Auricle Publishing Inc, 17141 Los Robles, Fountain View, CA. 92708-2027. The annual subscription fee for Audio Perfectionist Journal TM is
$98 per year for 4 quarterly issues. International subscribers please add US $20. Please remit payment in U.S. funds only. Audio Perfectionist Journal TM
accepts no advertising and is published solely for the benefit of its subscribers. Copyright© 2003 by Auricle Publishing Inc. All rights reserved. No part
of this newsletter may be reproduced in any form or incorporated into any information retrieval system without the written permission of the copyright
holder. Please forward all subscription requests, comments, questions and other inquiries to the above address or contact the publisher at
[email protected] Opinions expressed in Audio Perfectionist JournalTM are not necessarily those of this publication. Mention of specific products, services or technical advice does not constitute an endorsement. Readers are advised to exercise extreme caution in handling electronic devices.
17141 Los Robles Circle, Fountain Va l l e y, CA. 92708
Was this manual useful for you? yes no
Thank you for your participation!

* Your assessment is very important for improving the work of artificial intelligence, which forms the content of this project

Download PDF