Gramophone Dreams #61: The Art of Cable: AudioQuest, Canare, Kondo & the HoloAudio Serene preamplifier Page 2

Then Woo Audio's 3ES preamp/electrostatic headphone amplifier arrived. Putting it in my system as a preamp radically changed my thinking about what a line-level preamplifier could be and should do. Its push-pull 6SN7 voltage-amplifier tube and push-pull 300B cathode-follower output put a shot of LSD-clarity and power-triode umph into every signal passing through it. After Jack Wu collected the 3ES, I sulked and sighed longingly for a month, wishing for a full-function linestage that would be at once vigorous and invisible.

Then, recently, when the Linn Audio LP12 Klimax record playing system arrived (see my review elsewhere in this issue), I found myself needing to connect two phono stages and two DACs to a single preamplifier. Prompted by need and curiosity, I unboxed the well-boxed HoloAudio Serene. I had intended to review the Serene later, in conjunction with a follow-up review of the HoloAudio May DAC, because Kitsuné HiFi's Tim Connor said it was designed by Jeff Zhu to "show off the May at its best."


Well, Tim was right, and I can swear in public now that the Serene gave the May a dose of sunlight and a shot of adrenalin. What Tim didn't tell me was that it would also add grace and liquidity to the sound of the Mola Mola Tambaqui and punch and bass authority to the Denafrips Terminator Plus. I was most impressed by how the KTE version of the Serene enhanced the spatial presentation of Sutherland Engineering's Little Loco phono stage, adding force and rhythmic nuance while peeling away a layer of shadow that had previously obscured the perimeters of the Loco's projected soundscape.

Much like HoloAudio's Spring 3 DAC, which I discussed in Gramophone Dreams #60, the KTE Serene brought greater clarity and widescreen illumination to recordings, enhancing my ability to observe the full soundfield and exposing more of the vibrancy of the signals passing through it. The Serene's not-subtle penchant for illumination and vibrancy distinguished it from the RP-7 in a way that made the Rogue (with stock tubes) seem dark, humid, and dynamically reserved. The Serene displayed the type of see-through, look-into transparency I'd been searching for, the kind that made me think I could see past it into the interconnect and into the guts of the dCS Bartók DAC behind it.

For the longest part of my life, I thought linestages needed only basic things, 0–10dB of gain and to be as silent and uncolored as possible. But the more preamps I audition, the more I realize that when they are designed just right, they expose more of the musical energy embedded in recordings. With the best linestages, music doesn't sound reproduced. It feels reanimated.


Unfortunately, most reasonably priced linestages employ op-amps smothered in global and nested feedback and energized by excessively regulated power supplies. These software-optimized designs make preamps measure remarkably well—but they oversharpen detail, overheighten contrast, reduce atmospherics, and flatten space. The worst budget preamps make natural recordings sound artificial and compressed and overprocessed recordings sound more annoying than they actually are.

My auditions of HoloAudio's KTE Serene preamp suggest that it will outmeasure its competition and still deliver all that beautiful, supple, music-enhancing energy that allows recordings to sound relaxed, natural, and unmolested. I believe the Serene does these things because designer Jeff Zhu has proven, with his pristine-measuring, organic-sounding DAC designs, that he knows what unmolested recordings sound like.

The Serene incorporates a fully balanced, discrete transistor topology with a buffered, relay-activated volume control and a selector switch that allows users to choose between two balanced (XLR) inputs and three single-ended (RCA) inputs. The Serene has one RCA and two XLR outputs, all volume-controlled. Build quality, parts quality, and circuit layout would be exceptional at 10 times the price.

The Serene pre comes in two versions. The standard version costs $2698 and includes the standard remote, Mundorf EVO Gold and HoloAudio capacitors, and HoloAudio's O-core power transformer.


The KTE version reviewed here costs $3098. It includes HoloAudio's matching RMT-2 CNC-machined aluminum remote control and features upgrades including a flat-wire O-core transformer and Audio Note Kaisei, Elna Silmic II, Nichicon Muse, and HoloAudio house-brand capacitors. There is also an optional RMT-3 remote, which will control the Serene and any HoloAudio DAC.

Because the HoloAudio Serene preamp is so supertransparent—quiet, vivid, and flat-out exciting to use—I have already made it my one-and-only reference preamp. When coupled with Pass Labs' XA25 power amplifier, it is the most invisible audio amplification I've ever used.

What's the word? ThunderBird
Everyone knows I worship at the altars of transparency and hardcore explicitness—but I only kneel when those altars are inside the Temple of Dionysus. Resolution without blood and breath is not accuracy. Without those pulsing, life-animating forces I know are captured on my favorite recordings, the music is dead.


Those life-breaths of music can be killed off or diminished anywhere along the audio-signal pathway. But as my auditions are teaching me, preamps and interconnects—those simple things we expect to be the most electrically and sonically benign—can be rife with wave-smearing, force-mitigating, transient-dulling electromagnetic storms.

With that in mind, I removed AudioQuest's modest Red River balanced interconnect ($199 for a 1m pair) and installed their considerably less-modest ThunderBird balanced interconnects ($3900 for a 1m pair; $2900 for the RCA version) between the dCS Bartók DAC and the HoloAudio Serene preamplifier. I was curious to see how differently this more expensive wire—$800 more than the preamplifier—might perform.

For a few years, AudioQuest has had a top-of-the-line "Mythical Creatures" series of speaker cables, AC power cables, and HDMI cables; now they are adding "Mythical Creatures" interconnects. The series features the abovementioned ThunderBird cables and the even pricier FireBird ($5500 for a 1m RCA pair, $6900 for XLR), and the "yep-that's-pricey" Dragon ($9500 for a 1m RCA pair, $11,900 for XLR). These new cables employ the complete arsenal from AudioQuest's current thinking about the most effective small-signal conductor and dielectric configuration.

Every time I tried to transcribe some of the technology behind Mythical Creatures, my eyes glazed over, so I am going to save words. You can learn about AudioQuest's engineering strategies on their excellent website.


When I swapped AudioQuest's Red River interconnect for the balanced ThunderBird, my system's sound changed so dramatically that I immediately realized that my previous impressions of dCS's Bartók DAC had been skewed by the effects of the interconnect I had been using. What I had perceived as the Bartók's slight grain and grayness disappeared almost completely with the ThunderBird wires, replaced by a bright, waterlike clarity and a feeling that a dam had burst and the sun had emerged from behind a winter cloud. When we pay extra for better materials and more thorough and insightful engineering, we sometimes are rewarded not just with greater clarity but also with more intense, forceful, and naturally expressed sound. I had not fully grasped this before.

I repeated the above experiment, this time swapping the modest Red River wires for AudioQuest's ThunderBird cables between HoloAudio's $3698 Spring 3 DAC and $3098 Serene preamplifier. The improvements in tone and clarity were even more dramatic than with the Bartók. The improvements in transparency, force, and dynamics were similar in quantity and quality to what I experienced when I replaced a generic black power cord with AudioQuest's $1000 Thunder AC cable between my Parasound Halo A21+ amplifier and the wall socket. In both cases, my reaction was the same: "Damn! I ain't never taking this out." As of today, I hope I never have to use the Bartók and Spring 3 DACs—or any other DACs—with interconnects of lower quality than these ThunderBirds. Why would I? With these fancy AudioQuests, every time I push Play, I notice instantly how the music comes out of the speakers: big, strong, and vivid, to the max.

Ergo, QED, obviously
These experiences forced me to recognize how much my budget audio cables had been obscuring my view, not only of the components I've reviewed but also of the recordings I cherish. And yes, once again, I discovered that changing 3' of wire can be the difference between boring and thrilling, plain and beautiful, hard and supple, gray and colorful.

Finally, I am reminded that the importance of smartly engineered cables for good sound is not something audiophiles foolishly choose to believe in. It is something we experience directly.

Footnote 1: HoloAudio. US distributor: KitsunéHiFi 19410 Highway 99, Suite A #366 Lynnwood, WA 98036. Web:

Footnote 2: AudioQuest: 2621 White Rd. Irvine, CA 92614. Tel: (949) 790-6000 Web:


Jack L's picture


Frankly, many "cable deniers" do hear the sonic difference among audio interconnects. It is the high costs of such cables relaive to the moderate costs of their systems that deter them from taking it seriously.

My skeptical ears found pure silver audio cables sound so much better than non-silver cables many years back: elegance, see-thru transparency & delicacy in widest dynamic range: from pianissisimo ppp to fortissisimmo fff that non-silver cables can't come close.

So I started design/building them years back thanks to my electrical power engineering background involving subtantial use of wires/cables. I don't need to wreck my banks to finance the exotic cable vendors.

One special feature of the 99.99% pure silver interconnects I design/built since day one: the music imaging & soundstaging do remain sharp, stable & intact even when blasting the music at very sound levels, say up to 100dBs.

Those oxygen-free pure copper with/without silver coating I tried just fail to keep the sound picture in shape at such high volumes playing same music.

Trust me. Try to play yr favourite music starting from low to very high sound volumes while visualizing how the imaging & soundstaging change.
At average low sound levels, even cheapie RadioShack type cables will sound OK.

Playing at high volumes is an acid test for any audio cables.

Knowledge is power of saving.

Listening with own ears is believing

Jack L

mcrushing's picture

Hello, Jack.

I'm a "cable believer" (or, more accurately, a "cable comparer-and-yes-they-sound-different-experiencer").

What I do deny - and as a DIYer you may agree - is the notion that a relationship exists between what a cable costs and the musical enjoyment it delivers. I'm sure Kondo cable sounds wonderful but design, materials and manufacturing - not the price - are the reasons.

I've been down the cable-DIY rabbit hole for some time as well, and in my comparisons between bulk stuff like Belden and Mogami I think it's less that one sounds *better* than the other, and more that each sounds *very different* from the other. The choice comes down to your personal taste, system and goals. (Belden 3079a, PVC outer stripped, teflon tape wrap w techflex jacket is my current IC weapon of choice.)

Silver is the next frontier for me, but even in the DIY arena, it gets pricey. Loved hearing your impressions of cables you've made... care to share any design info?

Thanks in advance!

Jack L's picture


Positive ! Why not sharing it with cable DIY friend like your goodself ?

FTI, Belden is for RF industrial applications, e.g. cable TV etc & NOT for audio. I would not touch it.

Yet you can borrow the idea pf Beldon's mininum loss for long distance RF transmission - Teflon dielectric used to insulate the electric conductors.

The dielectric constant of Teflon (PTFE/FEP) is 2.1, highest among common plastics vs vacuum=1, & air=1.0006.

(1) signal conductor: Technically, silver is the metal of highest electrical & thermal conductivity on this planet. That's why I experimented it SONICALLY & have chosen 99.99% pure silver as the conductor metal for ALL my audio interconnects & power cords. I love the silver sound. IMO, it blows away sonnically oxygen-free copper wires with/without silver coating (common 'hi-tech' conductors used by many exotic cable makers. High purity silver is not cheap to buy at all. Yet yours truly managed to get such pure silver affordably. Lucky me !

(2) conductor dielectric (insulation) :
air is the best (virtually very very close to vaccuum) & is freely
available. Teflon is the 2nd best next to air, expensive though.
So I've chose BOTH for for silver wire insulation strictly on
technical ground.

(3) shielding against RFI/EMI noises:-

It is a must considering WiFi makes the RF noise environmemnt worse.

BUT I would not use any overall metallic shielding like most most,
if not all, brandname cable makers, including Belden.
Technically, any metallc shielding around conductor(s) will generate
interconductor capacitance which messes up the music quality, IMO.
My skeptical ears detect it.
I've succeed provided the neccessary RF shielding WITHOUT using any
external conductor shielding. It frees up the music big bigtime vs
metallic shielding, IMO.

(4) Protective outer jacket for the silver cable - not too rigid to provide flexibility & mobility of the cable.

Well, that's it. Any cable fabrication details that makes or breaks it, would be my little intellectual property which is close to the public.

Listening with onw ears is believing

Jack L

mcrushing's picture

A twisted or braided (i.e., Kimber) solution is all my non-engineer mind comes up with to "shield" ICs without a braided sheath (assuming balanced ICs aren't an option). I was hoping to learn your build technique, but keep your secrets if you must. I found a couple 99.99 silver/PTFE options (and one with cotton) at prices reasonable enough for an experiment, so I'll give it a try.

Don't count out the bulk stuff, tho. Belden 8402 and Mogami W2549 are for pro audio and please my ears more than entry-Audioquest stuff I once had. Yes, 3079a is for industrial automation, but it's a shielded, twisted pair of solid copper conductors in FEP "foam" with a spec capacitance of 8.5 pF/foot. Replacing the PVC jacket with teflon tape/techflex *might* improve sound and definitely improves flexibility. With cheap switchcraft RCAs and Cardas solder, I think it helps analog gear sound bigger and more 3D, with a bit more HF sparkle than the microphone cables above. It does have a bit less midrange presence, though.

IMHO, cables are a bit like wine. If you can afford to drink Grand Cru Burgundy, you absolutely should. But if that's not an option, there's a well-made $28 California bottle that'll please you just fine. You just have to look hard enough.

Jack L's picture


Yes, there are supplies of 4nines silver/PTFE jacket, pretty affordable.

But not my choice for the best sound. I only go for both PTFE + air. Air is the best insulation dielectric next to vacuum (1.00006 vs 1). So I built it my special way !

"Belden 8402 and Mogami W2549 are for pro audio"

What "pro audio" ? For miles-runs audio interface in auditoriums/recording studios ?? I would not use such "pro audio" cables for home audios. Besides, they are not pure silvers, not good enough for me.

"cheap switchcraft RCAs and Cardas solder"

Yes, no need of exotic brandname termination connectors, which are more money pitching than sonic effect, IMO. That said, I only use non-magnetic jacks which deliver the eligant silver-like music I want !

Which Cardas solder? I always use lead-free silver solder for best jointing. NO lead inside to damage our health !

"there's a well-made $28 California bottle that'll please you just fine."

Yes, for wines, I go for red wine only on occasons, e.g. X-mas parties as I am non-alcoholic. So no expensive wine for me.

The best low-cost red wine that gives me monthful of lingering rich, smooth & CREAMY favour is: La Crema brewed in Russian Valley. California. Wow, it tastes like 10 times its $40 price !!! Crema Red is not available in any wine stores, & is fully booked by restaurants due to its low yield.

One time, I spotted some La Crema white in Costco, San Jose during my visit to my younger son & family there years back. But no where I could locate any La Crema red there ! Shame !

No, unlike breverages & food, I want to get the best sounding audios for my money value by DIYing them right. No compromise !

Knowledge is the power to save !

Listening with own ears is believing

Jack L

Jack L's picture

La Crema wines brewed in Russian River Valley, CA.


mcrushing's picture

I believe Nordost traps air gaps beneath their PTFE insulation, while Tara Labs literally seals conductors within a vacuum.

Both seem far beyond my skills to manufacture.

I did run across Tempo Electric, who offer silver in "oversized" PTFE tubing on the idea minimizing the conductor's surface contact with the tube. To maintain this they recommend twisting your +/- pairs no more than once per inch – which seems problematic in an unshielded cable. Also: Silver + Air = tarnish. Silver sulfide is less conductive than 9999 silver. And, since RF is much more subject to skin effect than audio, a tarnished silver cable would seem present RF with a higher resistance. Am I right about that?

Btw, my 3079a ICs use "foamed" FEP. I understand FEP and PTFE to have a similar dielectric constants, and the foam structure adds air to the final product the jacket is made from. As I said, the spec is 8.5 pF a foot, roughly what I measure with my multimeter. On paper, PROFIBUS cable shouldn't sound good in a hifi system. Mogami microphone cable shouldn't sound as good as Audioquest Golden Gate. But they do to me.

As always, YMMV. But to anyone about to spend $60/foot for entry-level commercial hifi stuff, TRY IT. Worst case, you'll waste $25 and a couple hours on a working stereo pair of ICs.

I suspect I'll like silver better, as you clearly do. But to bring it full circle, I dropped the wine analogy as a way of saying I don't judge anyone's tastes or priorities. If I could afford to drink Grands Échezeaux or drop $20k like Herb's friend did, I would. But like you, I think DIY hifi delivers BIG value. So I appreciate the knowledge you shared. Cheers.

(And incidentally, I'm not into the creamy wines. I like less extracted, more bracingly acidic wines with only a little oak influence. Different strokes, different folks.)

Jack L's picture


Loosely twisted signal conductors NEED overall shielding to fight RFI/EMI, like Tempo Electric or whichever else cables makers do. So interconductor capactance will be added as bonus, affecting the music. The issue is that few brandname cable makers worry about the added sound affecting capactance. Either they don't detect the sonic defference or just take it for granted.

"Tara Labs literally seals conductors within a vacuum"

Yes, 4nines & purer silvers easily get tarnished by the sulpur contamination inside admospheric airs. Tara Lab uses vacuum to seal off the admospheric air. That's why such topnotch Tara Lab silver cables cost a fortune to acquire.

Well, I mangaged to home-brew my cable sealing which works fine after many years now: consistent ELEGANT muscial sound since day one till today. So it works. Of course, this very critical part of silver cable fabrication is not intended for DIY beginers.

"Mogami microphone cable"

I would not use any mic cable as interconnects. When you open up one such mic cable & check, you will know why.

Just imagine how much money I would have spent if I were to purchase pure silver interconnects (9 pairs for my rig today) & power cords from cable vendors assuming such 4nine silver cables & power cords available.

Work out yr own way that makes you happy! Not many audio guys get so addicated to silve sound like yours truly.

Listening with own ears is believing

Jack L

hb72's picture

you may use any other material for shielding as long as it is conductive (at least to some extent), as otherwise it is transparent or inert to electromagnetic waves (from WIFI or whatever).
Using e.g. carbon loaded plastics (as e.g. Audioquest does since ages), will help reducing susceptibility to EMI but since it is somewhat conductive, I wonder why such a material would not influence capacitance to some extent?

Jack L's picture


Sorry, any currents passing thru any conductive shieldings will generate a magnetioc field which interacts with the magnetic fields generated by the signal currents passinng thru the signal conductors - causing interconductor capacitance.

It just messes up the music sound, IMO.

The best is NO overall conductive shielding !

Jack L

Glotz's picture

Want the Serene more than blood itself!

s10sondek's picture

HR writes:

"Unfortunately, most reasonably priced linestages employ op-amps smothered in global and nested feedback and energized by excessively regulated power supplies. These software-optimized designs make preamps measure remarkably well—but they oversharpen detail, overheighten contrast, reduce atmospherics, and flatten space."

So how should I think about the tradeoff between the simplicity of circuit topology enabled by a well-designed op-amp versus the relative circuit complexity of a fully-discrete design?

Typically, subjective sharpening of detail is accomplished by injection of distortion or frequency response aberrations. But a great Opamp is measurably better in both regards. So how should I think about the mechanism by which an opamp causes this sharpening (or spatial flattening, contrast heightening or contrast reduction)?

In my experience, fully-discrete designs tend to blunt transient detail compared to an elegant opamp implementation. So I am trying to reconcile that empirical knowledge against a theoretical argument that is being suggested otherwise in this article. Is there a reasonable explanation that can be provided as to why a really good opamp should sound worse than a discrete component implimentation?

Also, I've always thought a well-regulated power supply was a good thing, reducing noise and distortion by ensuring that voltage remained steady throughout all current demands and with low noise, over all thermal and environmental conditions. Why would I ever think the opposite was true? Is there a tradeoff there as well that I should be thinking about?

I am not trying to be a jerk here. I am just trying to reconcile the posture of the physicist who opens the narrative against the statements that come later in the preamplifier discussion. And I don't think the correct answer is that "the physicist was wrong and learned his lesson." Rather, we should always be asking, why is empirical observation sometimes at odds with our scientific intuition -- and how do we reconcile the two?

Jack L's picture


An operation amplifier (or opamp in short) was a historic electronic device first invented by Bell Labs in 1941 using vacuum tubes. Not until 1958, Texas Instruments invented it in form of an integrted circuit or 'chip'.

In simple language it is a Dirct-Coupled (DC) high gain electronic voltage amplifier with a pair of differential inputs (one non-inverting & one inverting) & usually one single-ended output.

Its commom application is signal conditioning, filtering, to perform mathematical operations: addition, substration, integration & differentiation. Hence it is named "operational amplfier:.

As a physisist, hopefully you have already check out the circuit of a typical Op-amp: usually built up quite a number of miniature discrete biplar junction devices (or transistors/FET etc) with local & global feedback loops to achieve its controlled high gain for its common mathematical applications.

Yes, using standard bench test signals, e.g. sinewves, squarewaves & triangular waves, an op-amp can delivered immaculate figures, like THD, Noise level, etc etc.

BUT BUT an audio amp is for processing everchanging dynamic music signals formed of high orders of harmonics, not static bench test signals of single frequencies that simple. It is the capacitance, resistance & inductance inherently existed in the junctions of the bipolar devices forming an opamp that affect the passage of the complex musical harmonic signals. It makes even much much worse is the many built-in local & global feedback loops - very music harmonics 'unfriendly' - phase/timing feedback distortion.

Standard static bench tests does not measure realtime harmonic music signals.

Thank goodness, our ears can detect such dynamic harmonic & phase distortion caused by an complex constructed op-amps to the music that standard bench tests fail to achieve !

Only God knows when effective bench test methodology will be available to test dynamic music signals.

Less active electronics in the music signal path better will be the music signals sound. That's why yours truly only go for triode tubes which carry true linear signal transfer property. This is physics.

Listenig with own ears is believing

Jack L

JHL's picture

...from one of the true greats. This is a spectacular highlighting of what matters in fine audio and only what matters in fine audio - enormous enthusiasm, peer-confirmation of real, musical sound; working successfully toward that bliss, honoring the mysterious, and teaching the ethos by example - definitively punctuated by one of the greatest closing lines ever in audio journalism.

This piece leaves the reader yearning for more. If this is one story about one encounter, imagine the possibilities!

Glotz's picture

Smashed it on every level this (and every) month!

What's more he took his own journey to the next destination with honesty and candor.

hb72's picture

wonderful article of enormous inner beauty - at least to an audiophile cable nerd in search for music without holes.

tomita's picture

Has the Holo Audio Serene been measured here at Stereophile?

tomita's picture

Really now, is there a reason for not publishing the measurements?

John Atkinson's picture
tomita wrote:
Really now, is there a reason for not publishing the measurements?

Other than on rare occasions, Stereophile doesn't measure the products that are covered in regular columns, like Gramophone Dreams.

John Atkinson
Technical Editor, Stereophile

tomita's picture

That's too bad! It's too good of a preamplifier to miss measuring it!