Sonore by Simple Design microRendu: An Audiophile Odyssey

While USB DACs have been widely embraced by computer audiophiles for their ease of use and excellent sound, there has been a price to pay for the benefits associated with connecting one’s computer to these DACs. Noise, part and parcel of any all-purpose computer, is the enemy of music reproduction. The computer noise transmitted from the USB cable to the DAC steals from the user the ultimate sonic potential he could be enjoying. In recent years we have seen a multitude of USB enhancement devices to reduce this noise in spite of most manufacturers employing galvanic isolation in their DACs. Other approaches have consisted of low powered Linux or Windows servers to eliminate the all purpose computer entirely. Sonore by Simple Design has come up with another solution to allow one to use his high-powered all-purpose computer, but yet experience low noise USB audio.

The microRendu
The microRendu is a tiny computer that has been designed to offer USB DACs an optimum link to their computer or NAS libraries via an Ethernet connection. Simple Design has brought together their insights with Small Green Computer and John Swenson to create a microcomputer that runs on a modified Linux operating system called the Sonic Orbiter with a built in a circuit to regenerate a new USB data signal. The microRendu has a proprietary printed circuit board that does away with non-essential components that don’t serve music reproduction.

Michael Lavorgna has done a fine job describing and discussing the microRendu in his recent AudioStream review. It is highly recommended reading as a preface to reading this review. I will be presenting my experiences with the microRendu and the associated products that helped to bring this component to life in my audio system.

The Ethernet Network
My ASUS G501 JW laptop is approximately 60 feet away from my ASUS RT-AC87U router making a directly wired Ethernet connection very difficult to accomplish. I decided to purchase an ASUS Dual-Band AC1900 Repeater Range Extender, Media Bridge, Access Point with USB 3.0 (RP-AC68U) to provide an Ethernet connection for the microRendu. After all, I had had excellent results with Roon and the HQPlayer with the ASUS laptop running the Roon server, or as Roon refers to it, as the “Core”. I also decided to contact Stephen Mejias from AudioQuest to see if they would loan me several of their Ethernet cables for this review. Stephen put me in contact with AudioQuest’s Steve Silberman who was quite adamant that a wireless connection for my laptop was not really the best way to achieve top-grade sound from the microRendu. He suggested a setup that provided electrical isolation for the network with a system similar to that recently described and successfully used by Michael Lavorgna, but including the Netgear switch.

  • 2x TP-LINK MC200CM Gigabit Media Converter, 1000Mbps RJ45 to 1000M multi-mode SC fiber, up to 550m/1800ft
  • Tripp Lite Duplex Multimode 62.5/125 Fiber Patch Cable (SC/SC), 2M (6-ft.)(N306-006)
  • NETGEAR ProSAFE GS108 8-Port Gigabit Desktop Switch
My Ethernet network connected to the ASUS RP-AC68U is run in media bridge mode with no wireless extension or access point functions.

Comparing this isolated setup with using just a wireless connection for the Asus and the wireless media bridge for the microRendu, I found the sound to only improve by a marginal amount. The problem was the 4 SMPS power supply wall warts for the Netgear switch, the 2 TP-Link Converters, and the ASUS media bridge. Connecting these SMPS power supplies to a Shunyata Research Hydra DPC-6 v2 made a considerable improvement. But I decided to replace these noisy AC/DC SMPS power supplies with less noisy linear power supplies. To accomplish this, I used 2 HDPlex 100 linear power supplies to power the 4 devices. I then plugged the HDPlex 100 into the Shunyata Hydra DPC-6 v2. The results were simply spectacular sounding! The improvement in the system’s sound using this isolated Ethernet setup with linear power supplies displayed a larger, more open, and cleaner sound compared to the wireless setup. It was definitely worth the extra effort.

In addition, going to an Ethernet connection for my computer as opposed to WiFi solved an intermittent dropout of the music I was experiencing. This also fixed an issue that I was having with the HQPlayer streaming to the microRendu. After about 30 minutes of play, the connection between the microRendu and the HQPlayer was disconnected. With the Ethernet setup, everything worked perfectly.

The AudioQuest Ethernet Cables
I had asked AudioQuest’s Stephen Mejias to send me any 2 of his Ethernet cables for use in this review. Well, Stephen wasn’t going to let me off the hook that easily. A big box arrived with 2 of each Ethernet model that AudioQuest manufactures. Included were the RG/E Ethernet Diamond ($1195.00), Vodka ($339.00), Cinnamon ($89.00), Forrest ($49.00), and Peal ($29.00). All prices were for 1.5-meter lengths.

Michael Lavorgna has previously reviewed these AudioQuest Ethenet cables at AudioStream finding them to make an ever-increasing difference and improvement as one ascends the product line.

My findings were similar to Michael’s results. Not only did these cables sound better to me than my Belkin CAT-6 Ethernet cables, but ascending the product line resulted in a bigger, quieter, more relaxed sound that was not difficult to hear. I should note that I had greater difficulty hearing differences in these AudioQuest Ethernet cables until I installed the isolated Ethernet setup with linear power supplies. The Diamond and Vodka cables also utilize premium Telegärtner Ethernet connectors as an additional plus in physical performance. For me, the “best buy” in terms of price and performance was the Vodka. But I had no difficulty hearing the superiority of the Diamond, so much so, that I purchased two of these premium Ethernet cables for my system.

Michael and Steve Silberman identified “critical” positions for use of these AudioQuest cables. The Asus computer and the microRendu benefited from these upgrade Ethernet cables. The effect was less apparent using them in the other positions, but I was able to hear small improvements.

Associated Computer Components Used In The Review
As previously mentioned, I used my Asus G501 JW laptop running Windows 10 Pro 64 bit to act as the server or Roon Core for the microRendu. The Asus G501 JW possesses an Intel Core i7 4720HQ 2.6 GHz processor with 16 GB RAM and a very fast PCE Express X4 SSD. This laptop has 3 USB 3.0 ports as well as a Thunderbolt port. The Asus laptop was plugged into a Shunyata Research Hydra DPC-6 v2 distribution center to firewall the noise generated by this computer from contaminating my AC line.

The Asus was placed on a Synergistic Research Tranquility Base UEF grounded with the Synergistic Research High Definition Ground Cable / Grounding Block as was the computer. Two 8 TB GRAID Thunderbolt drives were connected; one for PCM and the other for DSD files. AudioQuest Coffee Thunderbolt cables were used. The GRAID Thunderbolt drives were powered by HDPlex 100w linear power supplies plugged into the Shunyata Hydra DPC-6 ver 2. The GRAID Thunderbolt drives and their HDPlex power supplies were placed on a Synergistic Research Tranquility Base.

I decided to use the MSB Technology Analog DAC with Analog Power Base with the new Premium Quad USB2 Module for this review. The Premium Quad Module represents MSB Technology’s most advanced implementation of USB for their DAC with superior isolation compared to their previous efforts. The Analog DAC was plugged into a Shunyata Research Triton v2 / Typhon using Shunyata’s Sigma Digital AC cable. The DAC was also placed on a Synergistic Research Tranquility Base UEF. The microRendu and its power supply sat on a Tranquility Base.

Using the midroRendu
The microRendu has numerous selectable outputs accessed by your web browser. I elected to use the RoonReady output and the HQPlayer NAA output. Although the microRendu comes with a small solid USB connector that is used between the microRendu and your DAC, I found it easier in my installation to use an AudioQuest .5-meter Diamond USB cable.

microRendu Power Supplies
The microRendu needs a 6-9 volt power supply that is not included. The microRendu is rated at a power input of 6-9 VDC at 1 Amp min continuous. I used three different DC power supplies for this review. The iFi Audio iPower , the HDPlex 100 watt linear power supply, and the Sonore Signature Power Supply. Sonore sells the iFi iPower for $50. I found this low-noise SMPS to work quite well with the microRendu. Its rated 9v output did run the microRendu warmer than the other 2 power supplies, but it was not an operating issue.

Using the HDPlex 100 watt linear power supply was a big upgrade sound wise from the iPower. This $399 linear power supply was able to supply 7 volts to the microRendu. With its independent 19V/12V, User Adjustable 5V-19.5V for the 9V output, and 5V output, this power supply was easy to use for this application as well as numerous others in my system. The HDPlex sounded bigger with stronger and better-defined bass compared to the iPower. I found the sound to be more open sounding with better definition compared to the iPower.

The Sonore Signature Power Supply at $1399 is Sonore’s all-out top grade power supply designed to drive the microRendu. Most everything about this power supply is deluxe from the Cardas-Sonore custom DC output cable; custom made footers with Sorbothane isolators, beautiful aluminum case, and many United States sourced parts. The Sonore Signature Power Supply was designed specifically to provide an ultra low noise, and ultra low impedance supply for the microRendu. The Sonore Signature has a custom ultra-low noise discrete linear regulator as well as the ability to provide the necessary peak current for the microRendu.

At first listening to the Sonore Signature Power Supply, I noticed a fuller sound than that of the HDPlex that was definitely more pleasing. But I had difficulty enjoying the over-all sound of the Sonore. The midrange and highs seemed to be recessed compared to what I was hearing with the HDPlex. I knew something wasn’t right, so I placed 3 Synergistic Research MIG 2.0 Isolation Footers under the Sonore Signature. The resultant sound was now magnificent with wonderful definition and balanced sound from top to bottom. The Sorbothane footers on the Signature might be just right for your installation, but substituting the MIG 2.0s made a world of difference for me.

The Sound of the microRendu
Setting up the microRendu to accept streaming from my ASUS running the Roon Server was extremely easy. I selected RoonReady in the microRendu software, and the microRendu then appeared in the network setting of my Roon server. It is just that easy.

I was very impressed with what I heard with the microRendu inserted in my system as opposed to playing my MSB Analog DAC directly from my ASUS computer. The noise floor dropped to the lowest I have yet experienced allowing an enormous soundstage to emerge that was holographic in its presentation. Fine details emerged, particularly with well-recorded orchestral music, which was previously obscured. The definition, focus, and micro dynamic qualities of the music were allowed to present themselves in a superior fashion compared to the ASUS with associated USB enhancement devices. The bass was tight and well defined with greater power and palpable dynamic changes. The biggest improvement I experienced was a lifting of a subtle veil and low-level grit that was not previously noticed until I compared the microRendu with the ASUS.

But as my editor Michael Lavorgna has previously stated, getting the small things right can make a big difference to the overall perceived sonic experience. My previous efforts in network isolation, linear power supplies, quality Ethernet cables, and the Synergistic Research MIG 2.0 footers allowed me to adequately hear what the microRendu was capable of. I then setup the microRendu software for the HQPlayer. I changed a few settings of the HQPlayer on my ASUS, and allowed the HQPlayer to stream to the microRendu. I converted everything to DSD128 for this exercise. The sound was excellent and worked perfectly with the MSB Analog DAC. I did find that the microRendu does not support native DSD with the Analog DAC making DSD 256 conversion unusable in this application.

I was curious to see how the microRendu handled the solo piano—an instrument that can be difficult to reproduce. I listened to the wonderful DSD recording of Narel Arghamanyan performing Rachmaninov; Morceaux de Fantasie; Etudes-Tableaus, and the Corelli Variations. Playing this recording through the microRendu resulted in the most convincing reproduction of piano overtones I have yet experienced. The dynamics of the recording were stunning with the full power and majesty of the piano reproduced in a convincing manner. The low-level background silence allowed the subtle venue acoustics to be easily identified.

Doug MacLeod’s reference recording Exactly Like This never sounded better to me when played with the microRendu. The focus and definition of the instruments and voice were exemplary. The microRendu allowed the music to flow effortlessly with outstanding dynamic life and detail. This 24/176.4 recording is recorded at a low level forcing one to turn up the volume. The microRendu allowed the transient quickness and impact of the drums to emerge from a deep black background.

An Interesting Finding
In Michael’s review of the Intel NUC and sonicTransporter computers, he felt that the sonicTransporter sounded better streaming to the microRendu than the Intel NUC. It would appear that the microRendu is sensitive to the computer streaming to it. Interesting enough, when I installed Fidelizer Pro 7.3 to the ASUS running Windows 10 Pro, I noticed an improvement to the sound that was easy to recognize.

An Audiophile Odyssey
When I decided to evaluate the microRendu, I never expected to get that involved in the Ethernet setup or power supply evaluation. But this audiophile journey allowed me to experience what the microRendu was truly capable of. I enthusiastically endorse the use of the mcroRendu to elicit exceptional sound from your USB DAC. The combination of the microRendu and the Sonore Signature Power Supply provided me with an outstanding musical experience that was truly impressive.

Associated Components Used in the Review
Ayre MX-R Twenty amps, Ayre KX-R Twenty preamp, Synergistic Research Atmosphere Level 4 interconnects and speaker cables, and Wilson Sasha speakers / Wilson Watch Dog II subwoofer.

Stephen Dupont's picture

Thanks for the review Steve. have you had any experience comparing a thunderbolt drive vs a USB 3.0 drive connected to your Asus? and, either of those to a standalone NAS, still using the Asus as a server?

Steven Plaskin's picture

Hi Stephen,

I'm afraid I haven't tried the USB 3.0 as my drives are just Thunderbolt 1.

I also don't have NAS.

Sorry Stephen.

les's picture

After shadowing the world of computer audio for years now (I'm still a CD-only listener), I can only come to the conclusion that perhaps the industry has gone down the wrong path - USB, that is. I can't believe the giration and the effort needed to make USB products sound their best. A whole cottage industry has developed around it, it seems. I wonder if another path - perhaps AES/EBU, ethernet, etc. - couldn't have been taken... All this is very fascinating to me.

Matias's picture

Indeed. Seems that before SPDIF and AES-EBU gave the quality but not the convenience. Then came computer and USB to open the convenience and file formats but not the quality. And now things are finally coming together as network and USB become better and better. Very interesting.

solarophile's picture

Nonsense. USB is fine and the asynchronous technology works well as far as I can tell.

I am happy that the industry has gone down the USB route and Gordon Rankin and others have done a grest job with pushing to the asynchronous standard. Never have I hear digital quality as it is these days and glad to be rid of CD transports. Just because there is this vocal "cottage industry" does not mean there are problems. Just that some people believe there are problems.

Michael Lavorgna's picture the chief designer behind the AudioQuest JitterBug, a device whose purpose is, in part, to reduce noise on the USB link.

Just because some people like you, who are skeptical without offering one shred of technical basis for your skepticism, i.e. "USB is fine and the asynchronous technology works well as far as I can tell.", doesn't discount, well, anything.

To speak further to your other posts here, I do not appreciate your needlessly derisive comments. If you have a question, ask. If you do not understand why someone says something, ask. Leave your unfounded nonsense at the door.

If you can abide by these simple requests, you are more than welcome to question away. If not, you are more than welcome to go away.

audiofool's picture

It's a marvelous device and so far its proving what I've always felt... there is no "free lunch" in signal processing. After trying both Foobar2000 and HQPlayer with upsampling to feed the microRendu , I find that upsampling really isn't necessary, sounds artificial

Priaptor's picture


Using my Diamond Plus IV which has recently been upgraded to Diamond V, I have to say, upsampling to DSD with HQP doesn't hit the high note for me. While Native DSD is still not supported with the microRendu for MSB, something I am hoping changes soon, feeding the MSB native non-upsampled sources to my ears is by far the best.

My initial listening to my MSB upsampled with HQP was "favorable" but ultimately failed as my "favorable" opinion proved to be "different" sounding and more artificial and changing back to JRiver either direct ASIO or using MinimServer via the mircoRendu, to my ears proved much better.

On the other hand, when using my T&A, HQP upsampled to DSD512 was the ticket.

I am not sure of how to extrapolate MSB's wonderful Analog to my Diamond V but for me upsampling to DSD using HQP doesn't cut it. Just my opinion.

Steven Plaskin's picture
Howie, Thanks for your input. You have a great DAC!
Priaptor's picture

The more I listen to MSB products and compare to others the more I appreciate them. Granted they are expensive but after over 3 years I still am extremely happy with it. I got the T&A for my headphone rig and despite the hype with HQP, while very good, is not on par with the MSB.

My point was I too experimented with HQP with upsampling to the MSB (didn't like it) and found the best way to use HQP with my MSB and the microRendu was to use the microRendu as NAA, use HQP as a player only without any upsampling and Roon as your library management. That combo gives me the best SQ. I would try that with your system and see how you like it.


solarophile's picture

Nice writeup Steve. But I don't get it... What kind of noise is it that you're worried about with the USB DACs? I've been listening to an Ayre QB-9 and it sounds great through USB and a NUC. I also don't see anything wrong with the USB DAC measurements by JA over the years. Should this not be verified by JA's objective results?

Steven Plaskin's picture

The fact is, DACs like your QB-9 are excellent sounding and Ayre has gone to great lengths to reduce noise from the computer getting into your DAC. But many of us have been finding that these low-powered computers sound better.

I guess you would have to try a microRendu to really know.

solarophile's picture

The issue I have is that when you attribute the difference to something like noise, you are claiming that these new low-powered computers are able to reduce something which is easily demonstrable with instrumentation which JA already can do for us.

Is there any evidence to show this is true before dropping >$600 without even a powersupply on something like this? It should be easily shown. Otherwise, I fear this "issue" of noise becomes another never-ending saga where every new product gets trumped up as yet another advance that some will claim they "hear" a "big" difference with "even lower noise".

Michael Lavorgna's picture
If you've read Steve's review, you'll have realized a number of things; due to Roon's architecture, the heavy processing is confined to a dedicated computer making Roon Ready device's processing requirements light weight, Steve is using opto-isolators, and on and on.

If you haven't done so already, I'd recommend reading "Open and Tolerant: MBL's Juergen Reis on Listening, Measurements, and (Un)Certainty". It *should* go a long way in helping you understand this subject more fully.

joneill's picture

Dr. Plaskin,

Thank you for the detailed review. Its a great read and appears the Sonore Micro Rendue is on the short list of possible replacements for my aging Macbook Pro.

In the interim, I've begun pulling Cat 6 wire to my equipment bunker and my router for a hardwire ethernet connection.

fritzg's picture

Sounds like this made the big difference in your listening experience and not the Rendu. This odyssey makes me wonder if Rendu is really only for high end systems with top of line cabling and power supplies for all associated equipment.

I'd love to see you take this exact set-up and replace the Rendu with a BeagleBone Black as your endpoint and see if you hear a difference.

Greyfossil's picture

I was with you until you got to the bit where you placed some mechanical isolators under a few capacitors, coils, transistors and a transformer in a regulated power supply and all of a sudden things were wonderful. The PSU should have been enough in splendid isolation but as assessed, it was not. "I placed 3 Synergistic Research MIG 2.0 Isolation Footers under the Sonore Signature". I'm really sorry and although I have experienced the difference some things can make, my level of curiousness (a non word to match my feelings about this) was stretched and my credulosity (another) spiraled into the incredulosospheer (another). Of course I cannot replicate what you have as I don't have such a sophisticated system but, and I'm not saying you made it up, my belief is now beggared.... ....... . Allowing for some effect on passive and active components through effective isolation is one thing; this is another.

I have the TP-Link and optical disconnection in place and am very happy. I do not have expensive feet under any static component although I have tried low cost sorbothane with no effect. I'm not an anti cable, anti tweek person really. It would be interesting to have your comments on why and therefore how such an improvement could be effected. In the meantime, sorry but but I'll stick with Michael's rational assessments of what he hears and equally importantly, why.

Steven, I really don't want to be dismissive or unkind but I'm afraid I have failed on this occasion. I just don't comprehend. There really is no need to respond. I'll live happily enough in my more limited cocoon knowing I cannot get to this level but comfortable with where I am.

I'll still read what you publish as I'm interested.

Steven Plaskin's picture

The MIG 2.0 feet are changing the mechanical resonance of the power supply. Ayre Acoustics recommends myrtle wood blocks be placed under their components. And yes, the overall sound of my solid aluminum Ayre components change.

I presently have Shunyata Research's 3 new Denali power conditioners in house for evaluation. Two of the passive power conditioners have similar guts, but are placed in different cases. One has a heavy base with isolation footers. The 2 conditioners sound different.

I do appreciate that you take the time to read my reviews and don't expect you to accept everything I write.

Thanks for your comments.


kukur9's picture

I was an unbeliever forever. I bought my Naim gear long, long ago and stopped reading most audio press. Sure I bought a DAC like everyone else and own a decent CD player but that was for convenience. (The LP12 is for "seriousness.") Audio marketing has always been better than audio engineering.

Like another poster, I listened to a lot of CDs too. And used the CD as a transport as well, feeding the DAC via coax. No audible difference between the two. I always, always felt like something was missing, and definitely inferior (sonically) to decent vinyl playback.

Then I somehow stumbled upon the Uptone Audio Regen, and suddenly the skies parted, god-beams shone down, and music was FUN AGAIN. To Steve's point, I thought maybe there is something to this noise and power business, because after all, that was part of the point behind Naim gear. So I thought what's another $100 and bought a Shunyata Venom USB cable.

Again, I was a skeptic and didn't expect much if anything. Well, I trust my ears. Another veil lifted and my humble office mini-system (Denon) was delivering Naim-like PRAT! I started doing long listening sessions, rediscovering my collection.

microRendu came next, and the rest is history. Or rather, maybe history is beginning if Ethernet based file delivery is next as it seems to be.

All this to say: I hope people put aside their need to know or rationalize or understand, and trust their ears. It's just audio, after all! (And electricity.)

Thanks Steve for helping us discover these occasional gems! Now, back to sticking my ears in sockets...

solarophile's picture

I don't know about the Regen thing, but your description makes me seriously doubt this kind of claim - "suddenly the skies parted, god-beams shone down, and music was FUN AGAIN". I mean, seriously, you're using the same DAC and did you now switch to some computer playback with USB? Surely that in itself compared to whatever CD you were using could have made a huge difference. A good asynchronous USB input itself is generally known to be objectively better than SPDIF coaxial (see JA's Stereophile measurements).

Then you say the USB cable switch lifted "another veil"!? I thought the skies were parted and god-beams already revealed itself!

Folks, it's literally descriptions and "testimony" as some have said like this that KILLS this hobby. Makes no sense and I think confuses and will disappoint most reasonable people.

Steven Plaskin's picture

Not much I can say to a guy who has made up his mind before trying these things.

And so it goes.

Michael Lavorgna's picture
Prove it. Point us "folks" to some studies that back up this wild claim of yours. You know, the kind of thing you demand from us, ;-), even though we have very clearly stated that AudioStream is not in a position to perform measurements.
solarophile's picture

Okay, let me be frank about my thoughts Steve and Michael.

I just turned 50 this year. I work as an electrical engineer. My main area of work is in wireless telecommunications. Signal to noise issues and jitter in RF and gigahertz communications are very real challenges I deal with every day. In the 2000s I was involved with setting up some major CDMA networks in the US and these days implement LTE solutions in this country and abroad.

I've been an audiophile for 40 years ever since my father introduced me to his Dynaco amps and Advent speakers as a child. I have been interested in computer audio for the last 10 years, thus the interest in the topics in this site. I typically use my 3-year old ultrabook to play to the Ayre DAC with Revel Salon2 speakers.

When Steve writes in the first paragraph:
"While USB DACs have been widely embraced by computer audiophiles for their ease of use and excellent sound, there has been a price to pay for the benefits associated with connecting one’s computer to these DACs. Noise, part and parcel of any all-purpose computer, is the enemy of music reproduction. The computer noise transmitted from the USB cable to the DAC steals from the user the ultimate sonic potential he could be enjoying. In recent years we have seen a multitude of USB enhancement devices to reduce this noise in spite of most manufacturers employing galvanic isolation in their DACs."

What do you think goes through my head as a professional engineer?
"What noise are they talking about?" Right?

The point is, after all these months with posts about the potential of noise whether it was with the JitterBug or now this microRendu, is there evidence that it even exists and that these products actually make a difference? If so where?

I have been a subscriber of Stereophile for the last 20 years and this is why I brought up JA as it would certainly be nice to undrstand what the readership is supposed to be trying to gain!

I believe doing so is extremely important. In my father's day (he was a computer engineer), professional friends would come and enjoy the sound system and talk about vacuum tubes vs. solid state, etc. These days, again frankly speaking, few fellow professionals would be able to get through an article like this without major concerns about the content. I believe educated, often professional music lovers with disposal income are the target audience for such devices. If this group questions the truth in articles and claims like this, then how is the hobby supposed to grow?

My apologies for using the word "KILLS" in the previous comment. No, I do not have demographic data for you, but I believe that if the hobby continues to build on myths (like this issue with USB being a significant problem even with good DAC?), and this is all the industry can talk about, then the hobby will continue its decline. A foundation for these claims appears to be absent.

Michael Lavorgna's picture
"The point is, after all these months with posts about the potential of noise whether it was with the JitterBug or now this microRendu, is there evidence that it even exists and that these products actually make a difference? If so where?"

If you do not know that noise is an issue in electronics, in networks, and in mixed signal systems, then I'd suggest you need to do some boning up on the facts. Our Industry Voice section (under Blogs), is a good place to begin. I already linked to the Juergen Reis article (did you read it?), but there are many others that discuss these topics in detail. I'd start with Juergen, John Swenson (also my review of the REGEN which includes a detailed description from the designer about the problems they address), the iFi article "USB Audio Gremlins Exposed: Beyond 1s and 0s", the interviews with Gordon Rankin, Charles Hanson, etc.

Once you digest that information, one of two things will happen; you will say, "Oh, I didn't know that", or "Prove that this noise is an issue with measurements". My guess is you'll fall into the latter category, which appears to be an affliction of some electrical engineers. My father was an MSEE, designed telecommunications equipment, and was an audiophile. I'm 55 and I've been talking to people who know what they are talking about for, oh, let's say 45 years. Never, in all of my discussions with my father and his EE/hi-fi friends, did I encounter the kind of narrow-minded attitude that is a product of thinking one knows more than one knows. Further, the most important aspect of hi-fi is music...and its enjoyment.

In terms of the relevance of your comments here, we listen to things and report on what we hear. We also share what the people behind the products have to say, refer back to the Industry Voice section.

The fact that you cannot accept what we write about what we experience, is something you need to come to terms with. Expecting a hi-fi review to answer all of your questions/skepticism, when you don't appear to know the basics of the issues at hand, is unreasonable, at best.

As I said to you already, if you have a question, ask. You managed to ask one question here, and I've given you a number of resources that will offer answers. Why you felt the need to couch that one question with all that other 'stuff', is a mystery to me.

solarophile's picture

Why so defensive and IMO rude? I'm certainly not looking for a fight, but bringing up that which I think is a concern probably to many technically minded readers.

I write fully so you can appreciate the concern in the context of where I come from and my own experience. I suspect there are many others who would be interested as well.

Yes, I have read the excellent piece from J Reis. I appreciate what iFi claims but it's the kind of thing advertisers write. And neither Gordon Rankin's nor Charles Hanson's interviews I have read address the technical reasoning in a fashion I feel can be automatically applied in an article like this one under discussion. Just because they discuss noise doesn't mean that the microRendu actually addresses what they commented on!

Can you do this for me as a longtime Stereophile subscriber?

Please ask JA if he can try his best to measure this issue around "noise" from computers since he has the experience and technical skills demonstrated over the years. I would love to see the microRendu improve the noise level from something like Steve's excellent MSB Analog DAC. It's the kind of topic which I think would make a great educational article in the pages of Stereophile.

Steven Plaskin's picture


Here is a response from an engineer consultant / friend of mine:

In a nutshell, here’s the thing of it.

As you know, digital converters (“DACs”) are clocked at a very high rate as part of the process. For simple old fashioned DACs like the TDA1541, that clock frequency is somewhere around a MHz when used in the famous Non-oversampling Mode. (NOS)..

That’s obviously outside of the audio band, but it’s part of the conversion process. (With modern chips that use over-sampling, it’s much higher).

Anyway, the basic mathematics of the process explain that noise that gets on the clock, or is injected at the digital chip level, will become part of the desired output. THAT’S NOT AUDIOFOOLERY - it’s true for every digital to analog or analog to digital signal process. PERIOD. Noise in the RF spectrum will become part of the audio spectrum easily enough. (Sad side note - a lot of this noise will also affect so-called analog products. Just how linear are amplifiers used in audio at a MHz or above? Answer - not very.)

Solarophile ought to go speak to the people who design the chips his systems use. They’ll tell him. Or, just read the data sheets and application notes these guys distribute.

Computer product manufacturers plain don’t care about digital or switching supply noise, as long as they pass the really, really lax FCC and other governmental requirements for emissions. It doesn’t affect their sales one bit. So, they do what they do.

CG's picture

This might be of interest:

It not only shows the effects of noisy power supplies, but also how hard it is to make worthwhile measurements.

Two comments from me:

1. In order to "see" very low level noise and other distortion products, general practice is to use averaging in the spectrum analyzer processing. The idea is that while the distortion products and the like are repetitive in nature, random noise is not, so averaging the results of multiple frequency sweeps will ignore the random stuff, but leave the products that repeat in each scan. That's great and all, but it hardly represents just what is presented to your ear on an digital audio sample by sample basis. It *might* be more productive to use a "Max" or "Peak" Hold sampling approach so that all the undesired crap is captured over a period of time. That's what you ear would be fed.

2. People say things like "well, look at that - it's at the -98 dB level - nobody can hear that!" and let it go at that. I guess that if that's your agenda, fine.

But, consider that this is in reference to the maximum signal level. Unless you like listening to pure sine waves at ear piercing levels, that's not how a home audio system is used.

Most people run the volume control so that the sound level is between 65 and 85 dB. (Some soon to be deaf people a few dB louder...) That's considerably down from the "where it clips" peak.

In addition, the total power of music or any other captured sounds is the sum of the powers of each individual sine wave that makes up the combined signal spectrum. So, no individual tone ever reaches the maximum level where clipping occurs and where the reference is stated.

That means that all these unwanted tones from noise and other interferers are actually much closer to the real individual sound tones than might be implied from the reference. That also assumes that they don't change much in level with actual program content.

Or, do they?

Much of the distortion due to non-linearity in analog processing circuits (amplifiers, etc.) is from intermodulation products. In a real system with lots of sine waves comprising the total signal spectrum, the IMD products can actually be a very significant portion of the total signal power. This was shown back in the 40's, if I recall right. This is why measurements of wireless and CATV systems, which have a lot of individual tones that make up the total signal, use measurements like ACPR and CSO/CTB to determine the performance of the system. These show IMD products that are far, far worse than simple harmonic distortion measurements would ever show. None of these measurements are that easy to make.

Anyway, intermodulation distortion is not very particular. Out of band signals will be just as happy to form distortion products within the audio band as outside of it. Note that these distortion products, unlike simple harmonic distortion, are only mathematically related to the desired audio signals and are not simple harmonic distortion. So, all they do is act like a form of noise, at least to your auditory system. By noise, I mean that in the most general sense - not hiss.

Most audio gear is not very linear outside of the audio band. Circuits that rely on high levels of feedback run out of open loop gain at some point as the frequency goes higher, so the circuit distortion performance can actually be dismal outside the audio band. That means that if some kind of out of band signal finds its way to these circuits, it's quite likely that IMD will form and may well be within the desired audio band.


Sorry for this reply being so long. It actually is a very long and complex subject and what I just wrote hardly even scratches the surface of the basics. (Yes, I've been working with this for a number of years. No, not in the music reproduction industry.)

Michael Lavorgna's picture
...has changed? If you started out communicating this way, we could have avoided a lot of unnecessary words. So, thank you for this more reasonable comment.

JA and I have discussed measurements for Audiostream, starting five years ago during my first interview for this job, and he simply does not have the time. As you can imagine, running the magazine, the website, reviewing, providing measurements for every review, and day-to-day admin and other tasks don't leave much time for a day off, let along additional work for our site.

Part of my frustration with your comments is you make assumptions. Your request for me to ask JA is just one example. Of course I've explored this possibility, so assuming I haven't strikes me the same as me asking if you remembered to put your shoes on before going to work.


kukur9's picture

You're right--I should've "restrained" myself and tried to channel my inner equipment reviewer. Maybe because I've done so (and do so most of the time) that I was just excited to share what I think is a game-changing step in this technology. It's effortless to turn on the audio now and hear... Well, I wanted to say magic but maybe it's emotion. It's just about impossible to put into words what I clearly hear that others seem to be puzzled by.

So, sorry about god-beams. How about this: Effortless pace, rhythm, and timing. Audible attack, sustain, and decay. Balance. Musicality.

Give it a try. Or rather, try to find things that don't require you to try hear so hard? (That's encouragement, not rancor.)

alphorn's picture

Hi Steven

Have you tried to connect the TP-Link dirctly to the Micro Rendu (without deviation via switch)?

Thanks for the instructive review!


Steve Bruzonsky's picture

I have used the TPLink optical converter, between a Linksys GB 1900AC router/access point and microrendu (using a CAPSv3 connected to Linksys for Roon core) and works like a charm! However, I just setup a Transporter i5 with Microrendu for a friend and the TPLink optical converter doesn't work in that system, in which my friend has a Centurylink GN router and a Netgear GB switch. Go figure?

Steven Plaskin's picture
Thanks Andy. I will be ordering some more optical converters and inexpensive regulated power supplies to get rid of the Neatgear switch as Michael has done. I'll report back.
alphorn's picture

I'm looking forward to hear about the findings of your (ongoing) Odyssey.

I can completely follow most of your results (especially the essential importance of power supply).

The advantages of optical links seem plausible. I think of connecting my router (in my cellar) via glass fibre to my audio gear (40m distance).

Thanks again for your upfront running!

vortecjr's picture

Steve, I appreciate your efforts in this review. Please allow me some follow up comments.

Optical fiber to Ethernet isolators can result in connection issues. Some users have reported connection issues resulting from firmware deficiencies on the devices and compatibility issues between the various components from different manufacturers. As such, we don't generally advocate the use of these devices. However, I understand that need for some to go "to infinity and beyond." That said, we are sponsoring Ted Brady (CA member) to develop a how to guide related to the use of these devices. The how to guide will show recommended setups and suggest components that work together well. I'll post a link when the guide is complete.

When using the HD-Plex power supply or any other multi-rail power supply care should be taken with powering multiple devices at the same time. The net result is that one could defeat the galvanic isolation of the components being powered from the same power supply.

I'm not sure is the AQ Ethernet cables tie the shields at both ends. Doing so could also defeat the galvanic isolation of the connected components. We typically recommend BlueJean cat 6a Ethernet cables.

For completeness we are in the process of updating the microRendu to allow native DSD into the MSB gear. Per MSB, "From there the DSD 1bit 2.8 MHz (DSD64x) and DSD 1bit 5.6 MHz (DSD128x) is mapped to our 1.4 MHz 24bit Ladder DAC with no analog filtering." DSD256 appears to be supported depending on the USB interface in elected for use with the DAC.

You said, "It would appear that the microRendu is sensitive to the computer streaming to it. Interesting enough, when I installed Fidelizer Pro 7.3 to the ASUS running Windows 10 Pro, I noticed an improvement to the sound that was easy to recognize." This is a generalization that can be misleading to readers. The RoonReady and NAA output mode on the microRendu receives streams from the player running on your computer. Fidelizer is affecting your player on your computer directly and indirectly affecting the microRendu. You can check this by using Fidelizer for local playback on your computer without the microRendu in the system. It's important to know that in these two output modes (RoonReady and NAA) the microRendu is just an output. If the quality of the stream from your computer is in fact improved by Fidelizer we would expect the microRendu to be resolved enough to present the improvement at its output. In contrast, when you use the DLNA/MPD or SqueezeLite output mode the microRendu is the player and software running on your computer is less of an issue.

Steven Plaskin's picture

There might be issues with the microRendu and the optical converters that Michael and I use.

Part of the possible issue with the micorRendu is that the basic processor used in it has a limitation of 800 Mbps for the Ethernet stack, which is also shared with other peripherals. Fortunately, these other peripherals aren’t active in an audio set-up. But, it still means that a 1000 Base-T to optical media converter, such as the one I use, might jam up the Ethernet portion of the microRendu. There is another TP-Link model that only works up to 100 Mbps, which presumably won’t cause problems with the microRendu.

Steven Plaskin's picture

Thanks for your comments Jesus

And yes, I did see that the Core computer running Roon and the HQPlayer did effect the sound I heard using the microRendu as you pointed out.

There are a number of ways to implement optical isolation for Ethernet and Michael has written a guide to his setup.

When using the HDPlex 100 to power the microRendu, it was not connected to any other components when I evaluated it.

Steven Plaskin's picture
I should have given the microRendu a Greatest Bits. My omission. I notified Michael.
les's picture


Would this be a viable implementation (for someone looking for an easy, least expensive microRendu setup)?

Laptop + JRiver (or the like) => crossover ethernet cable => microRendu => USB => DAC

I realize this will not be optimal, but it would be a starting point with much room for further improvement down the line. Also, can anyone comment on the efficacy of JRemote, for quick browsing of the catalog?

Any suggestions, clarifications would be much appreciated. Thanks in advance.

Steven Plaskin's picture
JRiver can act as a DLNA server. You would have to use something like Linn Kazoo to browse your library.
SoundsGood's picture

The green LED on my microRendu is on solid without the fiber modules inline and off solid when I have the fiber modules in line. The amber LED blinks with or without the FO inline as normal. I use TP-Link MC100cm modules. Any comment why this happens?
Seems Sonore has a concern about using FO isolation.

Steven Plaskin's picture
The green light indicates a Gigabit Ethernet connection. The TP-Link MC100CM is a 1 0/100Mbps converter.
SoundsGood's picture

The same Ethernet cable with FO modules gives a solid green LED on my OPPO.
So, no green LED on the mRendu and a solid green LED on the OPPO?
I do not have Gigabit Ethernet in my network.

Steven Plaskin's picture

From John Swenson:

"The amber light is the Ethernet activity light, the green light means you have a Gigabit connection. If the green light is off you are connected at 100Mbit."

Danny Z's picture

Hi Steven,

Thanks for your review and for your awesome website. I really enjoy reading your reviews and Michael's .
Just to let you know that i'm playing music from my QNAP NAS connected With Cinamon AQ cables through 1 Gig switch to the Network Renderer of the Analog Dac (With Power base), and the sound is just fantastic! You should try it yourself one day :)


austinpop's picture


Thanks for your very interesting article. It's motivated me to try the optical isolation method in my system. However, I did have a question.

As I studied your system topology, it seemed to me that the main purpose of what you were doing, and hearing, was galvanic isolation of noise prior to reaching your micro Rendu. So why not just put the optical isolators between your Netgear switch and the micro Rendu? Why put it so far upstream? By moving the isolators immediately upstream, everything upstream of the isolator is, well, isolated. So only the 2nd TP-Link FMC (fiber media controller) in the chain should need an LPS, and the only Audioquest Ethernet cable to worry about would the link from this FMC to the micro Rendu.

The reason I ask this is that my home network is a fairly complex collection of cable modem, router, WAP, and switches, and there is no cost effective way to put LPSes and expensive power conditioners on everything. That would be like boiling the ocean.

Amazon-willing, I will be able to try this, ahem, isolated experiment in a few days time!

austinpop's picture


I just read Michael Lavorgna's excellent article on this

I think he's confirming exactly what I wrote - that in most "normal" systems, it makes sense to put the isolators just upstream of the micro Rendu.

Presumably your rationale for putting it further upstream was the same as his - i.e. he has many streamer endpoints?

Steven Plaskin's picture
You are right about most systems. I tried isolating the microRendu but heard no irovement over the previous setup. I think the dirrence in play here is the Shunyata Research power conditioning stack I use.
Steven Plaskin's picture
Sorry for the terrible spelling. I just can't type with my iPhone :-)
alphorn's picture

Hi Steven

I have just ordered a device called GISO by a German company named artistic fidelity. Purpose of the GISO is galvanic isolation and filtering.

The CEO Mr. Koschnicke holds a Master degree in physics and owns a recording company well known for HD / 24 bit productions. The audio hardware products now sold under the artistic fidelity brand are side products respectively developments out of the studio.

Artistic Fidelity is hitting the market with several new products right now (website is under construction at this time). Beside the GISO (I've been ordering the GB for 300 Euros since the cheaper DS is reported not to run with Auralic Aries) they offer an ultimate PC-Audio USB interface called Afi+USB for 1'000 Euros. Furthermore they come up with AD- and DA-converters.

I will check and report

Steven Plaskin's picture

Hi Andy,

Thanks for the heads-up on this product.


alphorn's picture

Hi Steven

I've just received my GISO GB (which stands for 1 GB data rate).

The result to my audio (Auralic Aries) is very positve. Sound is more resolved and more natural in a way LAN-cables can't do. There is just more freedom.

What I like a lot is the fact that the GISO is a completely passive little box (means no mess with additional devices).


Steven Plaskin's picture

Thanks Andy. I will have to check this out.


alphorn's picture

Hi Steven

Mr. Koschnicke has 20 years of experience with ethernet/RJ45 connections (and digital recording/audio).

Perhaps you ask him what products he suggests for your system :)


stevebythebay's picture

As I'm about to finally turn in my Berkeley Ref DAC for an upgrade I ventured, based on a recommendation from Chris C., to get an Audio Alchemy DDP-1 to cover the multi-week "outage". I figured after I put the updated DAC back in, I'd move the DDP-1 into my home theater. Well that was all well and good. But Chris, Michael, you and others have been so high on the microRendu I figured I'd give it a go. Been a nice adventure in better sonics for the last few weeks. Much better connecting the USB out from my Mac Mini (late 2014 2.8 ghz / 8gb memory - running Roon Core and HQPlayer) with USB attached WD drive into the DDP-1, even with the "lowly" iFi power supply.

So, awaiting the Sonore Signature power supply, getting better Ethernet cables, employing a Cisco switch with the Sonore and Mac Mini, along with one of the 3 Eero mesh network access points to alleviate any noise from the Eero. I'm connecting the microRendu with a short Curious Cable.

When I've got all the pieces together, I'll lift this collection of gear (not including the DDP-1) and drop it into my audio system to find out how well it stacks up to my current Aurender N10 with Berkeley DAC.

One thing I was wondering about your testing was why, given your extensive use and reviews of USB cable in the past, you wound up using the AQ Diamond. Did you try any other cables and simply found this one the best performer, or was it more to do with a desire for a very short (.5 meter) connection.

Your network seems unduly complex, but I understand your reasoning. My approach hopes that a potentially electrically noisy Mac Mini and external hard drive is mitigated by keeping everything a wired network. If you or Michael have any suggestions/pointers to consider I'm game. Like you, my cable modem is far, far away from my audio system and am loathe to using WiFi, Powerline, etc. except to access the Roon server via its iPad app.

Steven Plaskin's picture

Hi Steve,

Nice to see you posting here. I'm looking forward to your impressions of the updated Berkeley.

I used the AudioQuest Diamond for the short length; but it does sound quite good. I would like to check out the WireWorld in the future. But you know how all of this goes!


stevebythebay's picture

In all the reviews of the microRendu it would appear that the USB connection wire choice is not either critical or very noticeable in the grand scheme of things. I'd simply thought with all of the testing you'd done, what with Synergistic, Wireworld, Shunyata, JCat, etc. this might have elicited a bit more scrutiny. I'll likely use the Galielo from the microRendu into the Berkeley Alpha USB, though I might see if a lesser cable makes any noticeable difference.

Steven Plaskin's picture
I did try different USB cables, but they were 1 meter. I believe different USB cables will sound different. The bass seems to be better with shorter lengths.
stevebythebay's picture

Interesting. I recall hearing that some technologies, like HDMI, are sensitive to length (something about wave related reflections) but didn't think USB was similarly sensitive. But maybe these two have that aspect in common. Maybe that explains why Sonore is offering up A to B type adapters, putting the microRendu as close to the receiver as possible. I'll have to experiment with varied lengths and types.

Steven Plaskin's picture
It has to do with the design of the microRendu. John Swenson discussed this at CA.
stevebythebay's picture

Guess I'll try the 200mm Curious Cable Regen Link, I've been using versus the Galileo when I move things into my audio system. I'd not recalled any discussion at CA re: short USB cables and the microRendu. You have the thread pointer to that discussion?

Steven Plaskin's picture


Check this out:

Quote Originally Posted by Confused View Post
Quite! I am trying to imagine how good a system, or indeed how good your ears and brain would need to be, to detect the difference between say a mR powered by a 'respectable' LPSU such as the Paul Hynes and a mR powered by the LPS-1. I would have thought the difference would be vanishingly small?
There is not a simple answer to this.

It breaks down into two pieces: how can anything in the upstream USB device (source) affect the sound quality that comes out of the DAC and how can the USB output of the microRendu be impacted by the power supply feeding it.

The honest answer to the first is "nobody really knows". There have been a lot of hypothesis floating around for a long time but with very little hard proof as to what is really going on. Part of the problem is that nobody knows what to look for since this is all about human perception, every time anybody tries to figure out exactly what is changing in the electrical output from the DAC to change the human perception nothing worthwhile comes of it. It seems human perception varies a lot from person to person, and from day to day for a specific person.

The result is that all we have to go on is vague correlations found from random testing. It's maddening from an engineering standpoint, but that is the way it is right now.

A bunch of experiments I did over a period of many years show correlation between signal integrity (SI) of the USB signal feeding the DAC and perceived sound quality. As the SI improves the signal sounds more "real". This has not been universally true for all people and all DACs, but has been true in a large percentage of the space (DACs and people).

Thus the purpose of the microRendu is to produce the highest USB SI I know how to do. SI consists of several things, jitter, noise on the signal, well formed signal (lack of over shoots ringing etc) and proper rise and fall times.

The power feeding the electronic circuitry is a very important part of achieving that high SI. As Barrows mentioned the noise on the power rails can cause increased jitter from the oscillator, and can directly wind up on the output signal.

The noise on the power rails comes in two parts: the inherent noise of the last regulator, (that is the noise when powering a fixed resistor) and the noise generated by fluctuating load current. This latter is particularly important for digital devices because the current drawn by the load (the digital circuitry) varies radically all the time. How the power supply behaves when given these changing load currents is very important. This is measured by an impedance VS frequency plot.

The Power Delivery Network (PDN) of the microRendu has been optimized to have very low impedance for high frequencies and mid frequencies, but not low frequencies. This was a deliberate engineering tradeoff, an important factor in USB SI is the length of the cable, shorter cables (everything else being equal) have better SI. So the device was designed to be very small so it could fit behind a DAC and connect with a very short cable. The tradeoff was that there is nowhere near enough room in such a package for enough capacitance to give good low frequency PDN impedance.

Thus for best performance of the microRendu the external supply must have very good low frequency impedance. Higher frequency impedance and inherent noise also make a difference but not as much as the low frequency impedance. A longer power cable has much less effect on SQ than a longer USB cable so keeping the case small so it could use a very short USB cable was the best compromise we could make.

Then there is an entirely different reason for the PS driving the microRendu making a differences in SQ: leakage currents. I don't have time to describe this in this post (I have put a lot in other posts here on CA) but in a nutshell it is noise that flows in a loop through the mains, power supply, DC from the PS, through interconnects (analog cables, USB cables etc) to another box and then back through the PS of THAT box to the mains. This is NOT, I repeat NOT the "noise on the AC line" that everybody talks about, it is completely different, it is hard to grasp what it is and how it works, but it is there with ALL power supplies that connect to the AC mains. Different supplies have different amount and spectrum of this noise and every system is going to be different as to where this current flows. It is usually a fairly low level effect, but once most of the other issues are dealt with it is frequently The largest contributor to sound degradation.

Thus differing supplies can have different leakage current affects on a system, completely irrespective to anything having to do with the first part of this post. Thus you may find out that a supply with higher noise but lower leakage, can sound better than the lower noise higher leakage supply.

One major problem with specifying leakage current is that it forms loops, (which is why it is called a ground loop) it takes at least two supplies connected to the mains to form this, thus it is impossible to specify the leakage current of a specific supply, it HAS to be in relationship with another supply. That makes defining a number for a specific supply extremely difficult. (If you have more than two AC supplies it gets even more complicated)

I know, way more than you ever wanted to know, but that is the state of things. The conclusion is that it is going to be extremely difficult to pick between two supplies based on published specs. You have to listen, and preferably in your system.

John S.

stevebythebay's picture

I have been listening to various USB connections between the microRendu and Berkeley Alpha USB. a) hard USB connector that came with the microRendu b) 200mm Curious Cable and c) Synergistic Research Galileo 1.5 M cable. The first two were about the same, sonically while the SR, despite its length, was best. It was less aggressive and had a lower noise floor. Just a more natural sound from vocals and instruments, with better placement in space and wider, taller, and deeper soundstage. In some recordings it was clear that the soloist in some songs is closer to the listener than in others. The differences between live and studio become all the more evident. And I didn't find any noticeable difference in the level of sonic output within the spectrum among the cables. With the SR everything was just more "relaxed" and far easier to listen to for extended periods. I suppose the way SR uses the active aspect of the cable mitigates against its length so SI is not compromised. But that's just a guess. Would be interesting to compare to a far shorter run of a Galileo if I had one to try.

steklo's picture

Hi Steven,

before your microRendu review you recommended the W4S Recovery very much. On CA someone recently wrote:

"The difference between stock PSU and LPS-1 powered Recovery, with Apple Powerbook as a source, was big when I tried. [....] and brought the quality very near a LPS-1 powered microRendu (in my opinion)."

Is there a chance to have you re-check that? If it would be true that a W4S Recovery with LPS-1 makes a MacBook sound nearly on par with a microRendu powered by LPS-1, well, that would be something!

Best regards