Audionet Max monoblock power amplifier

Three years ago, when I first heard Audionet's Max monoblock power amplifiers, I described their pairing with YG Acoustics Hailey loudspeakers "an absolute winner" and "definitely one of the finer systems at T.H.E. Show Newport Beach." At subsequent audio shows, no fewer than four other Stereophile Contributing Editors enthused about different pairings of YG loudspeakers with Audionet amplification. Herb Reichert, at the 2014 Rocky Mountain Audio Fest: "Everything had a kind of just right quality. Totally impressive!" Sasha Matson, at RMAF 2015: "Marvelous" on vocals, "rockin' and tight" on bass and drums. John Atkinson, at T.H.E. Show 2016: "Duke Ellington's classic Jazz Party in Stereo . . . was reproduced with terrific dynamics." Larry Greenhill at the 2017 Consumer Electronics Show on the sound in the YG Acoustics room, which included Audionet's Max monoblocks: "the sound on playback of Jake Shimabukuro's ukulele wizardry came closer to matching his live performance than in most rooms."

I visited the same room later during CES 2017, and wrote that it produced a huge soundstage, "with the low frequencies remarkably fleshed out and layered. Everything was perfectly controlled, with the speed, power, and weight of bass response astounding. Highs, however, were a bit recessed, and lacking in brilliance, and strings briefly turned wiry on the track's loudest passage" (footnote 1). What we didn't know from these auditions was how the Maxes might sound with other gear. A review would answer that question.

The Audionet Max ($30,500/pair), which was designed and are manufactured in Germany, costs almost the same as that of another German monoblock, AVM's Ovation MA8.2 ($29,900/pair), which I reviewed in the April 2017 issue.

The Max is based on what Audionet calls an Ultra-Linear-Amplifier (ULA) topology, originally developed for medical applications. The company claims that the technology "delivers results that nudge the boundaries of what is feasible with current measuring technologies. Even under conditions of maximum strain or in other borderline situations, any incidence of signal impurity remains almost undetectable, while the outstanding feedback damping ensures that the loudspeakers crisply perform to the limits of their capacity." I wonder what JA's measurements will have to say about that.

The Max has four separate power supplies—one each for the controller and input/driver stages, and one each for the output section's positive and negative half-cycles. The supply for the input/driver stage has its own 80VA toroidal transformer, with separate secondary windings for positive and negative voltages, while each of the two output-section supplies has its own very hefty 1000VA toroidal transformer—one for positive voltages, the other for negative. The controller stage—which is powered by its own toroidal transformer—monitors temperature, DC offset, and other parameters, and controls all amplifier functions. Should a fault occur, it is announced on the front-panel display, after which the Max disconnects from the mains supply. The Max's FET-based input stage is configured as a cascaded and bootstrapped differential amplifier. The output stage is equipped with eight hand-selected MOSFETs, with actively controlled bias current. For all that, the Maxes didn't run hot—the pair of them didn't heat up my well-insulated, 20' by 16' by 9' listening room, as do the Pass XA200.8 monoblocks ($42,000/pair) that usually drive my Wilson Alexia speakers.


On the rear panel are balanced and unbalanced inputs—the former are gold-plated XLRs of unnamed provenance, the latter are gold-plated, Teflon-insulated Furutechs—and two pairs of Furutech rhodium-plated speaker terminals (for biwiring). Also on the rear panel are a 15-amp power-cord input and, right above that, a small rocker power switch. The power receptacle, power switch, and speaker outputs are all so close together that, if both your power cord's input termination and speaker-cable spades are wide and your fingers are thick, you may have trouble accessing the power switch. I'm sure there's a reason for such cramped ergonomics, but it's not user friendly.

Equally disturbing, the positive and negative speaker terminals are close together. Bill Parish of distributor GTT Audio advised me to insert a piece of Styrofoam between my Nordost Odin 2 speaker cables' spade terminations to ensure that they would never touch. I made sure the lugs were tight, and checked them often.

The Max's Spartan front panel is far better designed. On it are an LED display, and a Power button that also controls a number of other functions. Once all inputs and outputs are connected and the mains power switch is turned on, the Max is in Standby mode. When you press the front-panel Power button, the message "Waking up . . ." is displayed. the polarity of the amp's AC cord be reversed—perish the thought—you're informed of that, too. Assuming all is in order, once the Max flashes which of its inputs is in use, you're ready to go.

Switching back to Standby triggers a sequence of messages: "Going to sleep . . . ," followed by "DISCHARGING NOW please wait . . ." When that message disappears, the Max is in Standby. If you change your mind and press Power once more, "RE-STARTING" is displayed. Parish recommended that I leave the Maxes on 24/7. Since they use very little electricity when no music is playing, and took quite a while to sound their best from turn-on, I followed his advice.


The Max offers several options each of operation and display: AutoStart, which permits timed operation; Auto On/Off, which turns the amp on from Standby whenever a signal is detected; and a warm-up countdown that enables the user to enter an amount of time, between 15 minutes and 99.45 hours, after which they intend to begin their listening. Also adjustable are the display's brightness and various other control functions. All of this is easily done by holding down the Power button longer than two seconds, then navigating the menu and making selections from it with short presses of Power.

If you leave the display on, you can choose to have it announce the input selected or the Max's temperature. Since the display text changes locations randomly every 12 seconds, I found it utterly distracting and switched off the display. That meant that, once the "Waking up . . .," etc., sequence was completed, the display went dark. Thank God.

"I have the time!"
When I first installed the Max monoblocks, Bill Parish urged me to replace the balanced (XLR) interconnects that I usually use with unbalanced (RCA). "For the balanced input Audionet uses an op-amp, which is just another piece of circuitry," he explained by e-mail. "While it is the best op-amp out there, you can still hear it. I don't want to make this a big thing, but I assure you that you will greatly prefer the RCA (or at least I do). BTW, I'll bet most of the equipment made today falls into this camp; they just don't know or tell you."

Given such strong advice, I switched to unbalanced interconnects, and stuck with them when I reinstalled my Pass amps for comparisons. While the Audionet Maxes don't require that shorting pins be installed in their balanced inputs when using their unbalanced inputs, the Passes do.


The first time I listened to the Maxes in my system, they sounded disappointing. I suspected there was some sort of incompatibility between them and all of my Nordost Odin cables—even though I've used Nordost Odin 2s with a wide variety of amps and speakers, and the sound has always been excellent.

I immediately contacted JA, and asked if it would be kosher if I borrowed a full set of Kubala-Sosna Elation! wires for this review. With his assent came a warning: I would need to establish, as a benchmark, the sound of the Pass/Kubala-Sosna combo before moving on to the Audionet/Kubala-Sosna, and "That will take a lot of time."

"I have the time!" Famous last words of Serinus the Idiot.

Footnote 1: The music was by Mason Bates, from his Works for Orchestra, with Michael Tilson Thomas conducting the San Francisco Symphony (SACD/CD, SFS Media 65). See
Idektron GmbH
US distributor: GTT Audio & Video
(908) 850-3092

Owen1's picture

I thought I was going to read a review of the Audionet MAXes but ended up feeling like I just read a rave review on the the Pass XA200.8 monoblocks and the Nordost Odin 2s at the expense of The MAXes and Kubala Sosna Elation!s. Perhaps the funniest or saddest comment depending on how you look at this was the admission that the MAXes may have been at a distinct disadvantage due to them NOT being on the Grand Prix Audio amp stands like the Passes WERE when evaluating. YOU THINK? That's like saying "we didn't feel the Ferrari handled as well as the Lamborghini but perhaps it was because the tires on the Ferrari were worn out retreads and the Lamborghini was test driven with Perelli P ZERO CORSAs". Wow!

Jason Victor Serinus's picture

I spent months with the Max monos in my system, properly positioned on the Grand Prix Audio amp stands. All my conclusions were based on that proper set-up, using the volume control of the dCS Rossini (which, I might add, needs no apologies).

The PNWAS visit took place after I had reached my own conclusions about the sound of Max monos. It was only during that brief visit, which extended over several hours, that the Max was at a disadvantage due to the lack of four amp stands and sufficient space to put them.

In retrospect, it was a mistake to have devoted so much space to comments made by others, at a demo where one set of amps was at a disadvantage. If you skip those comments, and read the rest of the review, I believe that it will make sense.

icorem's picture

I'm a proud owner of the Audionet Max.
I didn't find in the "associated equipment" section the pre that was used in that review.
I, as Owen1, found the review a little bit confusing...

John Atkinson's picture
icorem wrote:
I didn't find in the "associated equipment" section the pre that was used in that review.

No preamplifier - the dCS DAC, with its own volume control, was connected directly to the Max amplifiers.

John Atkinson
Editor, Stereophile

chhue's picture

Some years ago, I wanted to upgrade from my Bryston 4B SST. Because I already owned a Audionet Preamp, I aimed for the Audionet Amp. But I got an Audionet Max to audition at home. Because I was in search for a little more warmth and colour I finally got the Pass XA 60.5. But I always missed the last bit of rhythm and tight bass and also being more "effectful", traits the Max provided in my system.

I can relate to this review, though it was never a real A/B comparisson, because I've got the Pass some weeks later than I had the Audionet Max at home. Reading the review here, was greatly in line with my own memories.

The Pass XA 200.8 is surely an improvement over the XA 60.5.

The Pass and Audionet are very good amps, but for me they aim at opposing tastes. So I would not recommend choosing between them in a short period of time or even try to call one better or worse. At least this is how I understood this review, even if there is a slight bias in favour for the pass.

In fact, I think there are many amplifiers in between these sonic signatures. Because I build my own speakers with an cross over running in my computer, I've switched to two NAD M27. Though I might miss some of the XA 60.5's strengths, I do not miss their tradeoffs.

tonykaz's picture

I feel your concern.

It could've gone the other way and this comment section would have the Pass owners griping.

As an Salon Owner using Electrocompaniet Gear, I was never able to find any Amplification that could out-perform it but nearly all Reviewed Amps were "said" to be superior, I sampled nearly all to find them lacking.

Customers listen with their Eyes and when it comes to Amps.

Customers eagerly "listen" to all pricy Wire Products, they will always hear differences.

Customers can (and easily do) hear differences in PreAmps, nooooooooo problem.

Audiophile Amplification is the most Psychotic component, it's 2+2= something different for each person.

All this makes me wonder : What is the most popular Audiophile Amplifiers ever? I'd guess it's the NAD 3020.

Tony in Michigan

Ukcolin99's picture

Wondering what version of Toy you have my copy has neither track you referenced. Thank you

Jason Victor Serinus's picture

The 24/48 download from HDTracks. I have not heard other versions. Whenever possible, I work with hi-rez files or SACD (or perhaps, in the future, MQA CD), because the sound is so much better.

Double Fun's picture

Found this on Facebook, maybe it'll help:

"The Audionet review in this issue is perplexing. We have had 16 reviews of Audionet systems in the US since becoming the distributor and all of those reviewers come up with a different conclusion.
We strongly recommend using amplifiers with a companion preamp or at least a preamp. We don’t believe a DAC going direct with a digital volume control that trims bits is ever superior to a proper preamplifier. Further if a system is filled with tweaks and you are going to evaluate a component we strongly encourage detweaking the system and then reoptimizing the system. Only then can you really tell what a particular component really sounds like.
Also note that when using a digital volume control and playing DSD files the DSD is converted to PCM for the volume adjustment.
Any system’s sound is a sum of its parts. The system in which the MAX amps were placed might have been the Audionet showing off other parts of the system rather than the actual sound of the MAX.
At the most recent Axpona 2017 where the MAX amplification was used Jason reported,
“proved once again that this was a perfect, synchronistic match of components.
Perfect match? How about "highs that were brilliant without being piercing, impressive bass (albeit nowhere near as overwhelmingly powerful as from the far more expensive Sonja XV system), and excellent timbres? The bottom end may not have been as focused and clear as it could have been—this was a hotel room set-up, after all—but the excellence of timbres made for a winning presentation. Reservations aside, I loved it."
It is these types of experiences of our setups that peaked Jason’s interest enough to give the Max a try in his system. We just wish the recommendation of a preamp was used and a bit more care in the setup addressed. A true apple to apple comparison would have been appreciated.
We are not quite sure why Audionet was chosen for this beta-test-experiment of amplifier comparisons for an Audio Society when according to Jason, “the Maxes were at a distinct disadvantage” and the system clearly optimized for his reference amplifiers. Not only did the end results end in differing conclusions but amounted to a severely flawed experiment. True A/B test should be on an apple to apples basis which this clearly was not.
We appreciate John Atkinson’s measurements and his conclusion that “Max performed very well on the bench, offering high powers with very low levels of distortion and noise”.
We would like to highlight and clarify a couple of things in the measurement section. The measurement spikes which are shown in Figure 3 have nothing to do with the 1 kHz Signal. These are 60 Hz components from the mains. The procedure which Max turned off after 42 minutes has no practical attempt for a speaker but this was stated later on. All shut downs from the safety circuit are correct.

supamark's picture

makes you think it's a digital volume control? That makes zero sense from a design standpoint. Also, from the Audiostream review:

"...The summed latch outputs drive the second portion of the Analog Board: a balanced, Class A mix amplifier and filter stage. The mix amplifier and filter stage removes any unwanted noise and switching artifacts created by the high-speed latch mapping, and provides a variable-voltage, very-low output impedance, balanced analogue output signal."

That variable voltage output signal sounds a lot like an analog volume control, eh? dCS doesn't seem to say where the volume control is in the signal chain in their lit, but analog (after DAC) is the only one that makes engineering sense for a component like this.

supamark's picture

seemed as much about cables as amps. I don't personally buy into the cable thing, though I'd like to see all copper connectors/wire from imput to output. sending the signal through multiple metals, each with different electrical properties, is going to cause far more harm than some teflon dilectric layer can possibly fix. was funny that the the "B" option was overwhelmingly preferred (even when the amps were switched)... makes you think about the validity of those "lemmie swap in our high priced cables after listening to yours" demos, hmmm?

Would have much preferred the space spent going on about cables be spent comparing the sound of the two (or more) amps (even better if you could use multiple speakers since that's the interaction that determines if the amp will work for someone).

Joonas Viinanen's picture

In a triode, the signal has to travel through many connections and pieces of wire, many of different materials and very different connection methods, mechanical contacts, welded, crimped and soldered. The signal comes to the socket where it is soldered to a lug. If you are lucky, the solder lug and connection element to the tube pin are made of a single piece of metal. Sometimes, especially in sockets for larger tubes there are two. From the socket there is a mechanical contact to the tube pin. In Octal tubes, the pin is actually a shell which contains the wire which goes through the glass, it is soldered to the pin inside. Also the pins of directly heated triodes like 2A3 and 300B which have UX4 bases are made like this. The wire which traverses through the glass is not selected based on good sound, not even by good electrical properties. Here the thermal and mechanical behaviour is important so that no cracks in the glass appear when the tube heats up and cools down. Sound wise this type of wire is probably at the same level as barbed wire used in fences. Let's move on with the signal path. Inside the glass, there is another piece of interconnecting metal between this wire and the vertical rods which hold the electrodes like plate, grid and cathode. The actual electrodes then are clamped or welded to these rods.

So on average there are about 5 pieces of different metal and 5 connection points through which the signal has to travel. And that for each electrode of the tube. Assuming triodes only this sums up to 15.

How much difference do you think cables make if this is the reality?

Anon's picture

What did you think about the Xpander?

Also, which PowerCell are using?

Thank you for any answers you can provide! :)