Positive Feedback Logo

From the Kipnis Studios (KSS): The View From The Top - An Interview with Tommy O’Brien of Digital Amplifier Company

03-31-2017 | By Jeremy Kipnis | Issue 90

Tommy O’Brien Digital Amplifier Company

Kipnis Studio Standard (KSS) - Stereo Loudspeaker & Amplifier Testing Array with (Left to Right) Ologe Model FIVE, Ceratec Effeqct Mk. IV, and Waterfall Audio Niagara, Hurricane, & Elora Evo Glass Loudspeakers being powered by the Golden Cherry Amps from Tommy O'Brien at Digital Amplifier Co. (plus Nero Barky von Schnauzer - my assistant). The tiny Golden Cherry Amps are on either side of the fireplace & subwoofer (400 watts / 4-Ohm)


"Class-D? It can't possibly sound anything like music!"
"Analog sound cannot be amplified digitally and come out sounding analog."
"How can anything so small and light possibly sound as good as a classic Krell, Levinson, or Threshold?"

Amplifiers... their job seems simple enough: take a small signal and increase its voltage (and current) to drive one or more loudspeakers. Yet ever since this simple proposition was first enacted as part of the creation of the Telegraph and later telephone system during the 19th century, amplification designs have abounded from very simple to extremely complex, and all offering very different performance capabilities at verifying size, cost, and reliability. And the technology that developed initially from vacuum tubes, through solid-state and integrated circuits, and now all the way through digital has been the butt of most high end consumers for sound quality. Although there is a good amplifier design for every application imaginable including sound, image, light… any method of communicating information really, devotees of sound have shunned Class-D amps… until recently.

With all this said, no two audiophiles will agree on what non-Class-D amplifier is the best or even what qualities of any amplifier are associated with the finest fidelity playback systems due to their engineering. In fact, engineers are equally varied in their praise or damnation of various circuit designs and their implementation over the last hundred plus years. I can tell you as both an audiophile and recording engineer that amplification, no matter where it appears in the recording and playback chain, is no less important than any other feature we can put a name to: microphone, speakers, room acoustics, wires (ah Ha), vibration control, and amplification-both small and large in stature. They all contribute to the sound we cherish hearing.

Having worked up close with a good cross-section of both analog and digital amps (and their designers) over my nearly five decades in the fields of sound recording and reviewing, it is my pleasure to report digital amplification, often confused and merely lumped into Class D (or X) with a bad history of sound quality, has reached a point where it's fidelity is the equal of and in this case surpasses most every other analog circuit of the same type, cost, power output, and size. Apparently, long time Class-D circuit designs exist both simple and complex that sound fantastic, but why some and not others? I felt it necessary to request an interview with the designer of a pair of amplifiers, known as the Golden Cherry, and which I've had under review here at Kipnis Studios for over year, to clarify many confusing issues regarding this subject and his contribution to state of the art in both DACs (reviewed HERE) and Amplification.

Tommy O’Brien Digital Amplifier Company

Tommy's Golden Cherry Amps in the stacked configuration in my TRINITY Cinema powering (Left to Right) the Ologe Model One, Five, and Ten Loudspeakers

INTERVIEW with Tommy O'Brien (CEO & Designer, Digital Amplifier Company, a.k.a. Cherry Amplifier®)

How did you get started in this crazy field?

I became interested in audio at a young age. My father was an audio enthusiast and excellent engineer. He had a side business rebuilding and replacing the guts of Dynaco and Hafler amplifiers. He built custom audio equipment for family members, too, including a very elaborate 8-track deck and a wonderful pair of speakers with isobaric woofers. He built many other things as well, like computers, and was referred to as the "Godfather of Cable" where he worked—Jerrold, which became General Instrument, which became Motorola. He took me to local high-end stores to listen to the stuff he couldn't afford. I loved the sound! When I was in grade school, I routinely disassembled various pieces of equipment, like cassette decks, preamps, and such, while he was at work, and put them back together before he returned home. As my father aged, he was able to buy some high-end gear, starting with a pair of Magnepan panel speakers. He bought Krell and other highly regarding amps, a Theta D/A, and various preamps such as Audio Research. He built some amps from scratch, designing them from discrete transistor circuits. He taught me how to bias a transistor before I became a teenager. We spent many hours discussing circuit design. In 6th grade, I made a cattle-prod circuit from a Radio Shack Electronics Kit. So you know, the kit was not designed to make such things. I convinced my math teacher to put her hand on the contacts, and I shocked her. She had me suspended from school and sent to "the van" where I was psychoanalyzed. They really did think I was a troubled child, almost electrocuting my math teacher. I laid low for a while after that, but kept building circuits, everything from radios to lighting displays.

I became very interested in computers at about 14, and my big birthday present was a Commodore VIC-20 computer. My father helped me design a box that filtered audio into bands (bandpass filters) and digitized the levels using the joystick port of my VIC-20. I wrote the software to convert the screen display into a spectrum analyzer, entered this in the science fair, and was the first freshman in the history of my school to win the science fair. Not just my category, but the entire fair! Also mixed into my teenage years was an affinity for voice synthesis. I built a system (hardware and software) to learn new words automatically and read paragraphs using allophone-based voice synthesis. I went on to build an image processing system using a Commodore 64 computer. My high school physics teacher provided a large monitor and a CCTV camera for my system. I built a wire-wrap board with more than 60 chips, designing every last gate of my video capture device. That was when I was 17. I won all kinds of awards for that, including the IEEE science fair project of the year.

Early in college, I was experimenting with conventional amplifier design (Class-A and Class-AB) in my dorm room. I had some borrowed equipment, like a scope, soldering iron, etc., and piles of parts scrapped from this and that. I started getting more and more into analog circuitry. At one point, I thought I might get into software, because I enjoyed programming very much, but mostly to drive custom hardware, but there's just something about analog. For a school project, I made a 1W Class-A audio amp, and I was shocked at how wonderful it sounded compared to the other student's amplifiers. I drove this really nice little sealed-box (Focal drivers) speaker I made (one of the pair, anyway), and it was louder than I imagined 1W would be. There was this older guy in my class that built a Class-D amp for the same assignment. It sounded like crap and self-destructed during the evaluation, but the concept intrigued me. An amplifier that uses transistors like digital switches. Pretty cool. So, I started designing a Class-D amp of my own.

I was then the lead engineer on a senior project team to develop a high quality Class-D audio amplifier. The three other members of the team were my good friends, all of us EE majors. We were successful and received an A for the project. This was a single channel amp putting out something like 40W. At one point, it melted a tweeter off the face of speaker, but eventually, it was stable and sounded pretty good. It was just the beginning.... I graduated Drexel University with an Electrical and Computer Engineering degree in 1992. My project team decided to attempt a commercially viable design, so we started a company, business cards and all. We asked Drexel to release any claims to the design since it was started as a school project, and they agreed. The company fizzled after a while. We were more into women and beer than anything else at the time, but eventually I told them I'd like to pursue the venture on my own. They were fine with that, and I partnered with another friend to build a high power prototype PWM amp. The amp was a success, and I formed Digital Amplifier Company.

What exactly is Class-D?

Class-D simply means the output stage devices are used as switches. That's it. Over the years, some companies have resorted to making-up their own class designations for their switching amplifiers, such as Class-T; these are not officially recognized. Class-D amplifiers are sometimes called "digital amplifiers". This is a widely known and used misnomer, but there is truth to the term because these devices do use digital signaling internally, even if the input and the output are BOTH analog. At some point inside, the audio signal is represented by a digital signal (or signals) of only two states: high or low, one or zero.

Can you give me a brief history of Class-D (amps based on switches and PWM)?

The first commercially available Class-D amplifiers were produced in the early 1960s. Back then, key electronic components were not well suited for decent sound quality or even high efficiency. Efficiency was the driving force behind switching amplifiers for several decades, until very recent history. Class-D was thought to be fine for utility purposes, like public address amps, or self-powered subwoofers. Early Class-D amplifiers had major noise problems—the kind that results in a very audible "hiss" from the speakers. They also exhibited high levels of distortion, especially compared to their "analog amp" counterparts. Even by the 1980s (the era of the CD), it was difficult to find engineers with Class-D design experience. At that time, Class-D was not well known in the audio community. But by the 90s, it seemed everybody and their brother was taking a shot at Class-D design. It was this new thing in audio, with great promise especially where power density was concerned, like car audio.

It was difficult enough just to get a Class-D amplifier to work reliably and have decent distortion and noise measurements. Many designers were discouraged by the compromises involved with using available parts. Some were simply discouraged by the hidden complexity involved in what seems like a simple concept. There are so many "yeah, but …" factors involved, like components that are extremely sensitive to stray capacitance or inductance. "It's just surprisingly difficult" was a common reaction to Class-D design. The goal seemed to be simply making a decent amplifier that wouldn't break in the field. Reliability issues plagued audio companies, especially in the professional market segment where high power was king. Good sonics were a pipe dream at the time, and Class-D started getting a reputation for mediocre or even terrible sound quality.

But toward the turn of the century, the focus shifted to "direct digital amplifiers", which was Digital Amp Co's new concentration after developing analog modulated Class-D for several years. This term is used to indicate that modulation occurs purely in the digital domain, using logic circuits to perform Pulse Width Modulation (PWM) of the input data (digital audio source). During this time, Digital Amp Co licensed designs to ST Microsystems for the use in Class-D processing chips. That was becoming a bright new audio semiconductor segment.

By the time 2000 came around, there was a commercially available "direct digital" Class-D amp: the Tact Millennium. It even varied the power supply voltage as a form of volume control, which required expensive circuitry for semi-decent THD+N performance. Market acceptance was not so great, as I recall. The price was high (approximately $10k USD), and this type of implementation became more a novelty show piece than a reference audio product. One very good thing came out of the exercise—some acceptance of Class-D in the high-end market.

At this time, Digital Amp Co turned its focus back to analog modulation and sonic performance. The quest for decent specifications already resulted in success, but this is when it became obvious that specifications and sonics don't always go hand-in-hand with Class-D, possibly more so than with non-Class-D.

With the next decade came readily available high performance Class-D in high-end audio. This was high performance in terms of measured specifications. Digital Amp Co began selling the Cherry Amplifier, a large, heavy (up to 40 lbs./channel) design that put out plenty of power, nearing 800W into 4Ω. It offered high SNR and very low distortion.

Tommy O’Brien Digital Amplifier Company

The Golden Cherry Monoblock amplifier by Tommy O'Brien is tiny and lightweight compared to most amps with it's power rating, 200 Watts - 8 Ohms, 400 Watts - 4 Ohms, 800 Watts - 2 Ohms

So what are the advantages of Class-D (PWM – high frequency switched on/off) over say Class A (transformed always on) or AB (transformed always on at low level, high level switches to reserve power momentarily)?

There are some advantages to Class-D that non-switching amplifiers don't have, such as the lack of zero-crossing distortion. How important is that? Well it's pretty much the reason why Class-A is used in audio, to avoid the distortion that Class-AB amplifiers (still the most common amplifier class) exhibit when the signal is near zero (lowest levels). It's ironic how Class-A and Class-D share this advantage while one is typically very efficient and the other is very much the opposite.

Efficiency is the amount of power delivered to the load (speaker) divided by the amount of power needed to drive the amplifier. The efficiency of Class-D, which is theoretically 100%, makes it easier to provide power to the amplifier since there's less required for a given output. Most Class-D amplifiers are between 85% and 95% efficient, in reality. Efficiency also means less size and weight. This class of amplifiers can also deliver very good transient response and great specifications, now surpassing that of historically high performance analog designs, in fact.

I believe Class-D can surpass old fashioned analog amplification in general sound quality when properly executed. Some Class-D amplifiers, especially those used in high volume, mass market applications, can also be very cost effective.

Tommy O’Brien Digital Amplifier Company

The unbelievable Phantom Devialet Wireless Speaker couldn't exist it weren't for Class-D amp circuits, otherwise it would be the size of a suitcase!

What components go into a Class-D Amplifier design?

There's a modulator, which creates the digital signal (or signals), and the output stage, which drives the speaker, usually through a passive output filter. The output stage can be half-bridge, full-bridge, or even multi-tier. The output filter also varies per design in the number of stages and type of components used. Most filter-less Class-D amplifiers are typically low power and are low fidelity chips.

There's also a power supply, which can be one voltage, a pair of opposite voltages (bi-polar), or even multiple voltages of mixed polarity. Some designs are very sensitive to power supply fluctuations and interactions while others are nearly immune as long as the supply "stays up" when the demand for high current arises.

The modulator can be made many different ways and it is typically driven from the input with at least one feedback source. There can be multiple feedback sources, such as one from before the output filter and one from after the output filter. The feedback (or feedbacks) can be in a digital or analog form. The input can also be in a digital or analog form. Feedback implementation has a lot to do with the difference in sonics from one Class-D amplifier to another.

Modulation can be thought of as the heart, or better yet, the brain, of a Class-D amplifier. The modulation technique used in old Class-D designs compared the audio signal to a fixed frequency triangle or sawtooth waveform (the "switching frequency"). The waveforms and comparison can be accomplished in the analog or digital domain. In the digital domain, a clock is often used to control logic circuits. In this case, the clock frequency is typically much higher than the switching frequency. The switching frequency is typically between 100kHz and 4MHz. Our Cherry Amplifiers use both clocked and "continuous time" digital circuits with a variable switching frequency (up to 2 MHz), allowing audio bandwidth over 100kHz. Newer Class-D designs typically use variable switching frequency techniques with a typical maximum of 500kHz.

Can you describe other variations in Class-D designs?

Some companies, out of the few remaining that actually develop their own technology, have come up with a "trick" to their particular Class-D implementation. There are usually patents on these techniques as well, so the design is somewhat viewable by the public. Usually, the level of detail in a patent is not adequate to recreate the actual circuit. This is due to specifics like component selection and circuit board layout. Class-D patent topic examples include power supply voltage compensation, various modulator types, and specific feedback arrangements.

Be aware, however, that there have been many failures along the way. What looks good on paper often has problems in real life execution. In many cases, there are sonic penalties. Sometimes, a practical matter makes implementation to-the-letter impossible, or ridiculously expensive, or inherently unreliable. Often, components in the (electronic) paper design aren't "build-able" without enormous compromises or parasitic behaviors. How about an ideal component used in simulations where a real life version would be the size of a house?

For purposes of this discussion, we consider feedback to be part of the modulation. Just as with Class-AB or Class-A amplifiers, Class-D can be implemented "open loop", meaning there is no feedback wrapped around the output stage. "Direct digital" Class-D typically operates in this fashion.

To sum up, there are 4 major sections to almost every Class-D amplifier ever made: Power Supply, Modulator, Output Stage, and Output Filter. Which is responsible for the sound you hear, or the "sonic character" of the amplifier? The easy answer is "all of them"!

So most Class-D amplifiers are based on Class-D Chips bought off the shelf (rather than discrete components)?

In the low-fi and mid-fi markets, integrated circuits are widely known as "chips". The semiconductor companies are into making devices that can be sold in the millions of pieces. Otherwise, costs get out of control. This motivation has a tendency to dumb down whatever quality or fidelity to only price and measurable performance. The bottom line here is that the semiconductor companies don't care much for the high-end audio market where quantities are typically low.

There are huge compromises involved in miniaturizing certain electronic components at the chip level. One example are capacitors, which take up lots of space on a chip if they have more than a few picofarads of capacitance. This means adjusting other parts of circuits to compensate. Another example are power transistors. They not only take up lots of space, but they are difficult to mix with low power devices on the same chip. The "geometry" of chips is used to indicate the type of part in terms of "minimum feature size".

It is important to know that devices like processors and such are made to pack as many transistors into a given chip area as possible. Power devices are made to handle large currents or voltages. These two factors work against each other such that it's difficult to mix low and high power circuits, together. There's a similar problem mixing analog and digital circuitry. In this case, keeping the "digital noise" out of sensitive analog circuits can be very challenging. Good performance Class-D requires low noise and high power in most cases. A discrete, component-level design can combine different types of chips in the circuit so there's no need to compromise based on what can be effectively placed on a single (monolithic) chip.

What are you, personally, looking for when designing a Class-D amplifier?

Simply put, how an amplifier handles the speaker makes all the difference. An amplifier with outstanding specifications will probably sound better than one with poor specifications. However, there comes a point, at the very tip of high fidelity, where the quest for great specifications starts to take a toll on sonic performance.

Most amplifier testing is done using a resistor as the "load", which is commonly used as the replacement for a speaker on the test bench. Since speakers vary greatly in their reactance (how they react to the amplifier's "command"), measuring an amplifier with a "typical speaker load" isn't practical. Neither is playing an amplifier at hundreds of watts into a speaker with test tones, even if you have an anechoic chamber.

Some Class-D designers are in a race to win the "spec war" by beating their competitor's specifications with lower distortion and noise particularly. Specifications can be used as a basic litmus test to see if the end product is acceptable (The Stereo Review / Julian Hirsch Effect), but that next level of quality determination must be about the sound. So, using bench measurements alone to determine an amplifier's sound is a faulty exercise.

Sometimes the resulting measurements are used to determine a designer's skill level. This is true if the measurements indicate simply poor performance. There is a performance level below which an amplifier is considered to be solidly "mid-fi" or "lo-fi". Let's say 100dB SNR and 0.1% THD+N as an example. If measurements are in the "green zone", they may be there because the designer chose the numbers over the sound.

It takes an accomplished designer to know what measurement-improving design modifications adversely affect sound quality when pushed too far. This requires many iterations of modification, listening, modification, listening, etc..

But many great sounding tube amps have those "mid – low" specifications yet apparently perform miracles of sound recreation; often using quite a bit of negative feedback (which seems counterintuitive).

The "miracles" you speak of are debatable.  Noise and distortion are sound quality enemies.  A design technique for improving bench measurements is to increase feedback. Feedback is a term used for error correction by mixing some of the output with the input to derive the "new input" that drives an open loop amplifier circuit. The term "negative feedback" refers to the inverse of the output (or input) is used in the summation. The feedback corrects for errors (in the final amplified output signal), but it cannot turn back time. This means that the error being corrected has already happened, and the output signal will then be "directed" to compensate in the future. The speed at which the error correction happens determines how much the error "lingers" before a future correction. A good analogy is the steering of a car where the audio signal is represented by the car's position over time. If the driver wishes to go in a straight line, but momentarily steers slightly to the right, the driver realizes this "error", then compensates by steering momentarily to the left. However, in the end, the car may have been driven straight on average, but there was a jog to the right then left. Now imagine there were 10 seconds between the steering error and the compensation. The driver would wind up off the road! The speed of the feedback matters as well as the amount of feedback.

Most Class-A and Class-AB amplifiers use feedback, but with Class-D, there's an additional option. Does the feedback signal come from before or after the output filter? And it's possible to use both before and after filter feedback as well. So, which is better? How much of each? Well, the filter causes a delay, sometimes severe, that limits the speed at which feedback can compensate for errors. This doesn't mean after-the-filter feedback can't be beneficial. It does mean added difficulty in implementing feedback "around the filter". The benefit of wrapping feedback around the output filter is that it can compensate for errors caused by the filter itself. Designs that do this beg the question, "Why is the output filter creating errors in the first place?". In general, the better the output filter, the less error it creates.

How do other manufacturers use feedback in their designs to correct specifications if not sound quality?

The difficulty in achieving good open loop performance is common with Class-D amplifiers. And due to this issue, other designers tend to use gobs of negative feedback in general to get low noise and distortion. During the 1980s, many amplifier manufacturers, such as Adcom, used enormous amounts of feedback to tame their Class-AB amplifiers. This created the "too much feedback" artifact, where there seems to be an after effect to certain sounds. The result was harshness and listener fatigue.

Feedback varies with the audio signal frequency and error compensation decreases as frequency goes up. An amplifier's bandwidth is affected by the type of feedback used. There's also the question of stability. Imagine the error compensation is "out paced" by the audio input. This can lead to oscillation, where the amplifier can be permanently damaged (speakers too) because the feedback is unable to react fast enough to correct the error. Due to the stability issue, complicated feedback networks are used to reduce the effect on high frequencies. The result is sloppy high frequency performance even though the specifications would indicate otherwise. This is also the reason some amplifiers become unstable at very low impedance loads. Ironically, some speakers have decreasing impedance at high frequencies, so you can get a quasi-instability that causes smeared imaging or brittle sounding cymbals.

Many books have been written about feedback compensation. Finesse is required of the designer to use feedback properly and with minimal sonic penalties in any audio amplifier design, regardless of the class designation.

Other Class-D Issues?

Most Class-D is AC coupled. One reason for this is that half-bridge Class-D amplifiers suffer from the "power supply pump effect". This means that ultra low frequencies will cause the power supply to increase in voltage until the amplifier destroys itself. Another reason for AC coupling in any amplifier design, half-bridge or full-bridge, including non-Class-D, is that high precision (high cost) components are required to keep DC output offset low (a few millivolts) and stable over time and temperature. Servo circuits used to cancel DC offset add cost and complexity. Skillful circuit design is needed to adequately address these issues. AC coupling limits low frequency response and adds phase shift. This can make bass sound sloppy or weak.

Getting high bandwidth from Class-D is very difficult due to component limitations and the quest for high efficiency. In the module world, efficiency is more important than with discrete designs, and it's often used as a selling point. Heat dissipation can become a non-issue, but higher efficiency usually means more feedback and its associated sonic downfall. Doing things right in Class-D has the tendency to be on the pricey side. In any case, many amplifier manufacturers have simply quit trying to design Class-D. Sometimes the R&D effort is just too high. Sometimes the hardware itself becomes too expensive to sell. And sometimes it's just easier to buy pre-fabricated modules and call it a day.

Will we reach a point soon where a good Class-D Module exists with both great sound and affordability?

You mean, "Why not just buy someone else's amplifier modules, stick them in a pretty box with a power supply, and call it a day?" Well, it's easy to make mistakes outside the module (heat handling, connections, grounding, power supplies, etc.) that cause sonic problems as well as reliability issues. Longer signal paths and more interconnects (internal to the end equipment manufacturer's product) are required to use a module, creating sonic compromises. There's just no getting around the laws of physics. So even if the module sounds great, it may sound poor in the final amplifier due to other limitations imposed on the design.

The amplifier market (like the add-on separate DAC market) is getting crowded most recently by "me too" amplifiers built using pre-fab modules. Amplifier modules are designed to win as many "slots" as possible. It's a "one size fits all" type thing. The result is less bang-for-the-buck.

So what exactly are you doing in your amplifiers (and DACs) that is different from everyone else?

Our amplifier designs focus on sonics. We make output stages and filters with excellent open-loop accuracy, so they don't require much feedback to "correct". Our designs use innovative control loops so they don't suffer from lack of speaker control at the upper registers. Our designs benefit from decades of experience. Cherry Amplifiers don't suffer from the ringing and overhang of other Class-D topologies. We are offering speed and bandwidth without the sonic penalties. We have converted many tube lovers to our take on Class-D because the kind of smoothness that tubes can provide sounds simply amazing when it doesn't come with the added noise and distortion of tubes.

End of Part 1 (Interview)

Now… if you held out and read all of that, you're really going to want to know how the Golden Cherry Amps ($6900 per pair) will make your speakers sound, right? Well I've put these amps up against a whole slew of tried and true favorites in all categories of cost, size, age, and reliability including dynamics and ribbons made over the last 30 years. In my next column (Part 2) I will tell you precisely how and why these GOLDEN CHERRY Digital Amp Co. gems may be the single most transparent, uncolored, and precise amplification experience you are going to want to treat yourself to as soon as possible.


Jeremy R. Kipnis is a Producer, Director, Tonmeister, and Impresario. His heritage includes four generations of musicians and conductors. Working for various record companies, including Columbia, RCA, Nonesuch, Decca, Chesky, and his own label, Epiphany Recordings Ltd, he has been responsible for more than 450 award winning albums, SACDs, LPs, and CDs. His passion for photography led him to study briefly with Ansel Adams and Youssef Karsh. And his love for movies and television led him to design and create the Ultimate Home Cinema in the world, known variously as the $6 Million Kipnis Studio Standard (KSS)™. In addition to creating new cutting edge ultra immersive audio & video suites, Kipnis researches and reports on many topics in his three monthly columns, as seen in The High Fidelity Report, Widescreen Review, and Positive Feedback magazines.

Kipnis Studios (KSS)™


Epiphany Recordings Ltd.

www.EpiphanyRecordingsLtd.com  (New album coming soon)

The Review System (at Kipnis Studios & Epiphany Recordings Ltd.)

(items in italics are on loan from the manufacturer)

DIGITAL-to-ANALOG decoders

  • PS Audio PowerWave DAC II (with Bridge Card)
  • Theta Digital Generation VIII DAC (Mk. 4 updates)
  • Ultra-Analog A/D D/A (Custom 960 kHz Pro Tools Recording System)
  • iFi Nano DSD & PCM Portable & Self Powered D/A Converter System w/ 3.5mm Headphone and SPDIF Output
  • iFi Micro iDSD 768 8x DSD DAC & Headphone Amp


  • Mark Levinson No. 51 DVD-A Transport
  • Mark Levinson No. 37 CD Transport
  • Macbook Pro 17" 2.5 GHz Intel Core i7 w/ 8 GB 1333 MHZ DDR3 / 10.8.3 OS
  • Sony PlayStation 3 (SACD Playback via HDMI w/no transcode)
  • Audirvana Plus 3 (Stand Alone Memory Music Player for MAC)
  • Pure Music 2 / Pure Vinyl Recorder (LP High-Rez Remastering System & Memory player for iTunes)
  • Twisted Wave Logic Express 9.2 Soundtrack Pro – Adobe Audition CS6 (Digital Audio Editor)
  • Audacity Sonic Visualizer – AudioLeak (Digital Audio Analyzers)


  • ELP Laser Turntable (Custom Line-Level Output Cards w/ RIAA Correction)
  • Pioneer LaserDisc CLD-99
  • Technics SP-15 / SME 3009 Mk. 3 Arm / Shure V15-MRx Cartridge
  • Pioneer RT-909 2- track & 4-track Stereo Reel-to-Reel Player


  • Melos SHA-Gold Tube Preamp / Headphone Amplifier
  • Audible Illusions Tube Preamp / Phono Stage
  • Ultimate Attenuators (x2) – (31-step Laser Matched Resistors w/ Penny & Giles) - Balanced & Unbalanced Versions


  • Crown Macro Reference (x2) – Class A Modified
  • Carver VTA20S Tube (x2) (Pair running in Stereo & Bridged Mono)
  • Mesa Boogie - BARON Stereo Tube Amplifier (with adjustable feedback and variable Triode-to-Pentode performance)
  • McIntosh MC2102 Stereo or Mono (x2) & MC2301 Monoblock (x2) Tube Amps
  • Mark Levinson No. 33H Monoblock (x2)
  • Digital Amplification Company – 60 Volt KING Maraschino Cherry Mono Blocks
  • iFi Audio Micro iCan - Dedicated Adjustable Headphone Amplifier with Spatial Holographic Adjustment Algorithm


  • CERATEC Effeqt Mk. IV Loudspeakers
  • Symdex Epsilon 3-Way Loudspeakers (Built in 1993 by Kevin Voecks and Leeland Wallace in Gloucester, Massachusetts - USA) w/ Subwoofer (self Powered 12"): Cambridge Soundworks (x4)
  • Ologe ZERO 2-way Loudspeakers
  • Ologe ONE 2-way Loudspeakers
  • Ologe FIVE 2-way Loudspeakers
  • Ologe TEN 3-way Loudspeakers
  • Snell Music & Cinema (M&C) Reference Full Range 4-way Tower Loudspeakers
  • Snell Music & Cinema (M&C) Reference SUB-1800 (18") Passive Subwoofer (x4)
  • Waterfall Audio VICTORIA Evo Glass Loudspeakers
  • Waterfall Audio ELORA Evo Glass Loudspeakers
  • Waterfall Audio HURRICANE Evo Glass Loudspeakers
  • Waterfall Audio HF-250 SUBWOOFER Loudspeaker Addition
  • Bryston Stereo Subwoofer Crossover (x2)


  • Skogrand 421 Markarian Speaker Cables & Interconnects (Balanced & Unbalanced)
  • LessLoss DFPC Power Cords (x5)
  • Cardas Golden V, Neutral, Micro-Twin, and Clear Cables, Interconnects, Power Cords, Digital Connections, Microphone Cables
  • AudioQuest Ruby Speaker Cables & Interconnects, Glass Optical and HDMI Cables
  • Monster Cables M1 & M1000 Speaker & Interconnect, Pro 500 Microphone Cables
  • Mapleshade 1st Generation Experimental Speaker, Interconnect, and Digital Wires
  • Straight Wire, Kimber, Goldmund, Nordost, Sony, Panasonic, Radio Shack, etc.
  • iFi Audio II Gemini Split USB Cable


  • Magic Hexa Vibration Control Feet - 4 feet (x2)
  • Solid Tech Rack of Silence w/ Feet of Silence & Discs of Silence (x5)
  • Auralex 4" Studio Wedge Foam 2' x 2' Panels (x6)
  • RPG Skycloud Diffuser (x4)
  • ASC Special SQUARE Tube Traps (x2)
  • DIRAC Live Room Calibration Kit
  • Equitech 100 KVa Balancing Transformer (x2 : separate Digital and Analog services)
  • iFi Audio iPurifier (USB Filter - Signal Regenerator)
Share on Reddit0Share on LinkedIn0Email this to someoneTweet about this on TwitterShare on Google+1Share on Facebook134Pin on Pinterest1

Get our Newsletter


Recent Discussions