Gramophone Dreams #39: HoloAudio May D/A processor & Intona USB isolator

In contrast to phono cartridges and analog tape recorders, digital audio converters distinguish themselves by the fact that they can be fashioned in an almost infinite number of ways. Therefore, the odds against two manufacturers' DACs or ADCs sounding exactly the same are extremely large.

Nevertheless, I've heard countless audiophiles say that "bits are bits" and that today's digital-to-analog converters sound mostly the same. Some go as far as to declare today's DACs blameless—neutral—when debating issues of audio-system sound quality.

In my view, such opinions deny the likelihood that widely varied methods for reclocking, format conversion, oversampling, interpolation, current-to-voltage conversion, and reconstruction filtering affect the sound character of the music files being rendered.

My experience suggests that many budget DACs do sound the same: slightly blunt, unsupple, noisy, processed. A lot of DACs, no matter the type or price, make music sound processed, as if all the spinach in my cream-of-spinach soup had been chopped by a machine into pieces of exactly the same size and shape. Once I noticed this blenderization effect, I declared it a coloration.

I didn't realize it existed—this coloration—until I experienced digital without it in several of my friend's systems, all of which use expensive R–2R (or other ladder-DAC) converters made by companies like Totaldac, Denafrips, or MSB. Listening to those systems prepared me for my discovery of HoloAudio's modestly priced Spring DAC at Capital Audiofest's CanMania in 2017. The minute I heard the Spring, I knew: This $1698–$2698 DAC was recovering music via some sort of silent, un–digital-sounding channel, just like my friend's DACs.

Three years later, I am auditioning HoloAudio's newest, most expensive converter, the $3798–$4998 "May," and wondering: could this DAC sound as natural as my friend's fancy DACs?


The HoloAudio May
According to the website (footnote 1), the new DAC's name, "May," refers to "the possibility of"—the word that means that—and not the second month of, well, "spring." I am not certain what that is meant to convey, but I do know that my transition from using HoloAudio's "Spring" as my daily driver reference DAC to using the May has opened my mind to the possibility of a more utopian digital future.

Tim Connor, of HoloAudio and KitsunéHiFi (they are the same company), calls this new, two-chassis converter "The ultra-endgame DAC." That is a pretty fantastic claim, but, based on the high levels of build and sound quality Jeff Zhu (HoloAudio's founder and chief engineer) packed into the reasonably priced Spring DAC, I believe that the May DAC is an assault on the state of the R–2R converter art.

The May DAC is a dual-mono, direct-coupled, R–2R resistor-ladder design. It is housed in a matching pair of thick-walled chassis made of black-anodized aluminum and polished copper. Its substantial (20lb) linear power supply is housed in a separate chassis, connected to the DAC-circuitry chassis by a flexible, well-constructed umbilical cord about a meter long.


Both chassis measure 17" × 12" × 2.2". Their official combined weight is 39.8lb. The May has a typical array of digital inputs: two coaxial, one optical, an AES/EBU, a USB, and two I2S over HDMI. It also has balanced (XLR) and single-ended (RCA) analog outputs. The May uses an op-amp input stage and a discrete, class-A transistor output stage.


The converter chassis runs class-A warm-to-the-touch, while the power supply chassis stays room temperature, suggesting that its resistors and four power transformers are over-spec'ed. Just after the May arrived, I tried it with the DAC on top of the power supply and heard no apparent noise. The next day, I moved the power supply chassis to a more strategic location, on the floor about 20" away.

Like the Spring, the May has been optimized for nonoversampling (NOS) use. However, it includes three front-panel–selectable oversampling modes. (Oversampling is accomplished via an AK4137 chip.) The first OS mode, labeled OS, "supersamples" PCM to higher frequency PCM and "oversamples" DSD to higher-frequency DSD. The second OS mode, designated OS PCM, converts all PCM and DSD inputs to oversampled PCM. The third OS mode is called OS DSD. It converts both PCM and DSD inputs to oversampled DSD.

The May does not include line-level preamp functions or a volume control. To my surprise, however, it did come with a heavy CNC-machined black-anodized aluminum (with copper buttons) remote control. This remote allows users to mute, change inputs, or turn off the display. There is also a Mode button, which allowed me to switch from NOS to oversampling modes—on the fly—from my listening position. I had much fun with this.

KitsunéHiFi offers the HoloAudio May DAC at three levels of parts quality:


Level 1 is the standard, two-chassis May costing $3798. It uses the same XMOS XU208 USB input module as the Spring2 and features dual O-type (not toroidal) power transformers wound with flat, 6N copper wire. The capacitors in the heavily regulated linear power supply are by Rubycon ZLH, Panasonic FC, Vishay, and Mundorf EVO silver/gold. The fuse is a Shurter Gold type.

Level 2, at $4298, adds a new, "enhanced" USB XMOS input module with the HoloAudio Titanis 2.0 circuit. KitsunéHiFi claims this new USB module is "twice as powerful/capable as the one used in the Spring2." It also adds "custom proprietary" 1000V 1µF HoloAudio branded power supply capacitors (replacing the standard Mundorf EVO silver/gold caps).


At $4998, the Level 3 KTE (Kitsuné Tuned Edition) reviewed here uses hand-selected DAC modules for "best measurements and dynamic range." After hand-selection, these DAC modules are covered with black CNC-machined copper shields emblazoned with the Kitsuné fox logo. The KTE edition replaces the standard copper hookup wire with 1.5mm pure OCC silver wire "soldered direct to the PCBs with Mundorf Supreme 9.5% silver solder." In this version, the standard Vishay caps are upgraded to KTE-brand capacitors, the standard IEC inlet connector is upgraded to a pure silver/rhodium plated Faston type, and the power fuse is upgraded to a Red Nano brand with graphene and quartz filtering materials and gold-over-silver contacts.

After reading the May KTE's parts inventory, I am sure the DAC (and cable) deniers among you are screaming "snake oil!" However, a more astute DAC consumer would have noticed that, typically, passive R–2R DACs lean heavily on specialty parts options while sigma-delta and FPGA DACs do not. Why do you think that is?

Consider this: Maybe, when the effects of oversampling, format conversion, interpolation, active signal processing, and reconstruction filters are minimized or eliminated, the sonic character of passive parts becomes easier to recognize, like low-level detail emerging from a lowered noise floor.

Intona USB 3.0 Superspeed Isolator & Impedance Controlled USB cable
Speaking of noise: I have experimented some with galvanic isolators. Off and on, I've used ifi Audio's $379 nano iGalvanic3.0, which I find does its prescribed job of quieting, and unfuzzing, music streamed from my computer. The sonic effect of this noise reduction was that recordings sounded sharper, cleaner, and more three-dimensional—but not always more enjoyable, so I've never felt inspired to keep any galvanic isolator in my system. My brain rationalizes: if this add-on isolation is really so necessary, why is it not included with every DAC?


Connor sent me the $329 made-in-Germany Intona Technology 7055-C USB 3.0 SuperSpeed Isolator, which he said should provide better sound with the May DAC, especially when I use my laptop as source. Of course, I wanted to experience that.

According to Daniel Stämmler, Intona's CEO and chief engineer, "The intended purpose is complete isolation of both data and power lines. [The] resulting sound quality improvement is the effect of breaking all the noise coming from the host computer."As best I can tell, the Intona isolator breaks ground loops, minimizes coupling between data lines, and protects USB ports from power surges and voltage spikes. Sounds helpful and harmless, no?

Footnote: 1 HoloAudio and Kitsune HiFi, 19410 Highway 99, Suite A #366, Lynnwood, WA 98036. Web:

MhtLion's picture

Many reviews/experiences I read elsewhere point to that May may be one of the best DACs for DSD. I strongly recommend using HQPlayer to oversample any DSD to DSD512. I believe using HQPlayer for OS is clearly better than using May's own OS function.

Bogolu Haranath's picture

Next HR could review the PrimaLuna EVO 100 tube DAC with tube clock ($3,000, reviewed by Hi-Fi News) :-) ........

Charles E Flynn's picture


I literally spent hours trying to find DAC whose performance could be improved with this device. None of the high-end DACs cared one way or the other. They continued to work just as well as they did without Intona. After much frustration, I pulled out my old and discontinued Schiit Modi 2 USB which I know is sensitive to USB (power) noise and managed to eek out a bit of difference. Here is the Modi 2 directly hooked up to my PC:

Bogolu Haranath's picture

ASR said essentially the same thing about AQ Jitterbug ...... Also, JA1 did not find any measurement differences with Intona or Jitterbug ...... See, footnote 2 :-) ........

Canuck in WNY's picture

I’m convinced that somehow Audioscience has found a team of reviewers and readers who are earless, as they never seem to listen to any equipment.

hb72's picture

On my DAC the AQ JB does a big difference in treble, also mids and bass. More analog if you want. Also ifi has excellent products that do similar things. Measurements that do not cover the essential improvement may be meaningless. As much as the proof of the pudding is in the eating and not in mass spectroscopy, the single most relevant test remains a long term listening test on your very own equipment. Here the JB and ifi’s purifier make a real difference.

Glotz's picture

Here is the more important section of the 'review' that the Earless Dude at ASR stated:

"What is that? You hear an improvement? Have a loved one put the isolator in and out of the loop of the USB connection 10 times. If you can tell reliably 8 out of 10 times and can document it in a video, I will pay you $350 so you can buy this product. Otherwise, it is not the spoon that is bending. It is you!"

Lol.. your EARS are lying to you now. Not that his tests are inadequate! YOU ARE. At least ASR is great comic relief.

"If you can't measure it, it doesn't exist!"

hb72's picture

That is indeed quite funny.

dc_bruce's picture

Still trying to get my brain around at column entitled "gramophone dreams" that concerns a DAC.
So, are measurements of DACs meaningless? Has Mr. Atkinson been wasting his time all these years, measuring digital stuff?

Jim Austin's picture

Stereophile has rarely measured equipment reviewed in columns. However, JA has performed measurements for a follow-up review. Those measurements will appear in the September issue.'

Jim Austin, Editor

Norbert2nd's picture

Dear Jim,
Having had a look at the measurements and diagrams I was really wondering if those and the listening have been made with the PLL enabled. If you look at the J-Test published on ASR there is one diagram titled "J-Test PLL On vs Off". JA's measured jitter diagram to me look more like the PLL Off curves.
Can you comment on this? The PLL setting has never been mentioned in any of the articles and due to the somewhat hidden access to configuration mode of the May-DAC it can easily get overlooked.
Thanks & best regards

John Atkinson's picture
Norbert2nd wrote:
Having had a look at the measurements and diagrams I was really wondering if those and the listening have been made with the PLL enabled. If you look at the J-Test published on ASR there is one diagram titled "J-Test PLL On vs Off". JA's measured jitter diagram to me look more like the PLL Off curves...

I looked at the ASR jitter measurements of the HoloAudio May - see - and the measurement with the PLL on looks superficially similar to mine.

In any case, I measured and listened to the May in the condition in which it had been auditioned by Herb Reichert. My measurements and my followup review will be posted to the Stereophile website in a week or so.

John Atkinson
Technical Editor, Stereophile

tonykaz's picture

Brilliant observation Mr.dc bruce.

Of course, the dollar costs involved in 33.3 are staggeringly outrageous, which sort of makes the constantly improved digital options our prefered direction over this next decade.

Its seems that we should've had another NEW music format by now! These Music Format systems don't have a long & useful life. ( We've been thru at least 7 competent and popular formats since the 1950s ).

I like 16/44.

Tony in Venice

Bogolu Haranath's picture

May be a new column called 'Digital Dreams'? ........ It has a nice ring to it :-) .......

Bogolu Haranath's picture

Other choices are 'Digital Planet' or 'Digital Corner' :-) .......

tonykaz's picture

There is a future out there, about to reveal itself.

I'm anxious about it.

It's kinda like we are about to witness Ben Franklin finally revealing the power of mother nature's electricity, after thousands of years waiting.

Something good is coming, I can feel it.

Tony in Venice

Bogolu Haranath's picture

We will have 3-D Holograms of musicians playing in our listening rooms :-) ......

tonykaz's picture

I'm looking for more knowledge and more access to accurate and reliable information in all things including our fatalities i.e. Cancers, Heart issues & Drug addictions.

I'm anticipating a significantly improved transportation system and implementation of Thomas Jefferson's Declaration of Independences famous promises.

What Society is doing now is the Prelude to Thomas Piketty's forecasting. ( along with the majority of the World's Economists )

Our little music systems will continue to improve, just like they always have.

Tony in Venice

tonykaz's picture

Who are you talking a-boot?

Tony in Venice

ps. Piketty is the French Economist Professor

tonykaz's picture

Tony in Venice

Jim Austin's picture

Let's delete the non-audio-related economic politics please.

Jim Austin, Editor

Bogolu Haranath's picture

Ok ...... I have deleted mine :-) ......

tonykaz's picture

Hello Mr.JA2,

Mr.Audiophiliac just presented a rather brilliant 'both sides argument' pertaining to Chinese Audio Gear in our USA Marketplace.

I think he nails the critical Service aspect of outfits not standing behind their products.

Steve G has been one of your greats.

Thanks for being there for us.

Tony in Venice

ps. for the record, China has been a quality leader in manufacturing for the last 500 years or so.

Jim Austin's picture

Stereophile is agnostic on place of manufacture. We do however try to provide information that will allow our readers to make their own decisions; information on place of manufacture can be found in the Specifications box in every formal Stereophile review.

Of course, customer support is crucial. To secure a Stereophile review, a company must demonstrate that they are able to provide adequate customer support to U.S. buyers, whether it's through a dealership or online. If a product cannot be repaired in the U.S., or shipped overseas and back quickly, then there should be a mechanism for rapid replacement.

Jim Austin, Editor

Bogolu Haranath's picture

Only a French economics professor can 'amass a wealth of gigantic' intelligence and knowledge :-) .......