CHORD ELECTRONICS DAVE

[Rob Watts]

I have finally gotten round to listen to the AQ jitterbug, as a quest to understand where the RF noise problems from the source was coming from - via the mains or the USB.
 
Firstly - very much - YMWV as RF noise is a funny thing and source noise may affect other parts of the system, so the findings here may well be different in other circumstances.
 
So I did the tests with just a Win 10 lap-top (jRiver), Dave and pair of Nighthawks, in a hotel in Singapore. The idea was to try to evaluate whether the SQ improvements I heard when disconnecting the lap-top from mains and running off battery was from the USB connection or via the mains supply. Now Dave is galvanically isolated from LF and RF noise, and this makes a huge difference in SQ. But the galvanic isolation is not perfect as there exists a 2pF coupling capacitance. Now you may think 2pF is nothing to worry about; but at  1 GHz it is nearly 80 ohms impedance, so RF noise in the GHz will couple through - and I know that GHz noise is significant in causing noise floor modulation and hence changing the sound quality.
 
I used my design lap-top, an MSI Intel i7 machine, as it will be noisier than my music lap-top (HP Pavilion) which is very power efficient. This is so it would be easier to hear any changes.
 
So adding the Audioquest Jitter Bug did indeed improve SQ - it was fairly easy to hear it, with it sounding smoother, warmer, with less sibilance and glare. Instrument separation and focus was better. These are exactly the kind of change I hear with lower RF noise, so it was not unexpected.
 
Next was to do the listening test with the lap-top on batteries - and then it became impossible to reliably hear the effect of the jitter bug. Additionally, the improvements heard were much smaller with the jitter bug connected when you disconnect the lap-top power. Here is a rough assessment of SQ performance with 100% for best, 0% for worst:
 
Lap-top battery power, no jitter bug       100%
Lap-top battery power with jitter bug      100%
Lap-top mains power with jitter bug         75%
Lap-top mains power no jitter bug              0%
 
So what does this tell us? Basically the most significant path of RF is residual coupling via the USB, not via the mains supply as the jitterbug alone gets you most of the way.
 
Use the jitter bug for mains powered USB sources - it is worth the small cost.
 
The best sound is still with a Win 10 lap-top on battery and for critical listening that is what I do. Battery operation and Dave's galvanic isolation means effectively perfect RF noise isolation from the source.
 
Note also Win 10 is technically better than iOS xx, Android or Linux as these driver less USB's do not guarantee bit perfect data at the DAC; but Chord's Win driver does resend faulty packets, so we can guarantee bit perfect data for Dave. It would be interesting to do a listening test of driver less against Win.
 
Rob

 

[romaz]

  I have finally gotten round to listen to the AQ jitterbug, as a quest to understand where the RF noise problems from the source was coming from - via the mains or the USB...

Thank you for your analysis, Rob.  This is very helpful and it further convinces me that music servers produce very stubborn amounts of RF noise, potentially much more than what is coming from the mains.
 
Based on your comments in the past, I had drawn the conclusion that the two biggest challenges for a DAC in dealing with a music server are (1) jitter and (2) RF noise.  My own ears have convinced me that you have solved the jitter issue.  Obviously, RF noise has been a tougher nut to crack even with the galvanic isolation that you employ but I wonder if combating it solely at the USB level is the key.
 
Over the past few months, as you know, I have tested a variety of sources and as I have reported with my DAVE, some USB sources sound better than other USB sources and I know others have reported hearing these differences also.  I have played around with various USB cables in addition to an AQ Jitterbug and indeed, there were differences which support what you are saying, that USB output is noisy and we would all do well to address it at this level.  
 
This is where I start to get confused, however.  As I have previously reported, the differences I heard using a cheap printer USB cable against a $1,200 Clarity "Natural" USB cable (which has been the best USB cable I have heard thus far) was maybe 5%.  To my ears, adding an AQ Jitterbug, didn't improve things much more.  Adding a USB Regen didn't improve things at all.  Could it be that there is only so much "cleansing" that you can do at the USB level?
 
After reading your post this evening, I went back to my WIndows laptop with SSD, my well-shielded Curious USB cable that has served me well, installed an AQ Jitterbug and ran it off its battery.  Using Roon (and nothing else, like HQ Player) and the Chord ASIO driver, I played a few of my reference tracks and confirmed playback was bit-perfect.
 
I then switched to the microRendu using its stock switching power supply connected straight to the wall and using a $10 generic USB cable, I compared it against my Windows laptop.  The microRendu has no free USB port to allow insertion of something like an AQ Jitterbug.  Just like my WIndows laptop, I used Roon exclusively and confirmed bit-perfect playback.  What is interesting, however, is that unlike my comparison of different USB cables where the best cables were only subtlely better than a $10 printer USB cable, the microRendu was clearly superior to the Windows laptop, even to my wife who doesn't have a well-trained ear, because the differences are not so subtle.  While the microRendu is smoother and has less glare (similar to what you described), even more noticeable is the considerably enhanced dynamic contrasts, better-defined bass, more incisive transitions and increased clarity.  Even in stock form, it outclasses any other server I have heard to date and its superiority is easily appreciated with the DAVE.  While my better power supply with a lower output impedance for the microRendu hasn't arrived yet, those who are using such a power supply with their microRendu are suggesting even greater improvement.    
 
As I have tried to understand the differences between the microRendu and other digital sources like a Windows laptop, the best way to describe it is that it is a "one trick pony."  Others have made this claim but the microRendu goes further than anything else I have seen.  Like many other computers, it contains a motherboard with CPU, RAM and an ethernet input as well as a USB output. It contains no local storage which can introduce noise.  Unlike the Windows laptops that you and I own, this motherboard is a completely custom design and contains none of the extra parts that are deemed unnecessary for music playback.  There is no graphics card, wifi chip or any superfluous ports.  Consequently, this motherboard is smaller than a deck of cards and so the signal path, from ethernet input to USB output is extremely short.  Furthermore, between the power to the ethernet PHY and the power to the USB subsystem are four ultra-low noise regulators, some of which have very high PSRR resulting in numerous and redundant levels of isolation.  With the microRendu, what I have found is that the USB output is so clean that an expensive USB cable with enhanced filtering capabilities adds no benefit.  It is so well isolated that the dirtiest of networks seem to have no impact on it.  What this suggests to me is that it is better to prevent noise than to have to remove it later and that possibly, once a signal is degraded by noise, it becomes irreversibly engrained in its fabric and there is no way to fully restore it.  Of course, this is my own speculation.
 
I wish that you could hear a microRendu for yourself and see if you agree.  

 

[stvc]

Not sure when is Rob leaving singapore, i do have a set 

.

 

[Rob Watts]

Leaving tomorrow!
 
But I will be getting hold of a microRendu to try - I am intrigued, particularly as the observation is that it is smoother, which is indicative of lower RF noise. Its a puzzle as battery supply with no earth connection ought to be the lowest way of transmitting RF noise into Dave. But weird things happen with RF, particularly at GHz frequencies. And it still may be nothing to do with Dave, but a symptom of other pick-up in the rest of the system. As I mentioned, YMWV!
 
Rob

 

[lovethatsound]

Chord Dave review on the what hifi site for those that are interested.

 

[ecwl]

http://www.computeraudiophile.com/content/698-sonore-microrendu-review-part-1/
 
I think the article above has the most information on the design of microRendu. Maybe Mr. Watts would comment more when he has had a read and a listen...
 
I'm really hoping I don't "have to" or get too "tempted to" upgrade my entire system to one based on microRendu...
 
Oh, well... The never-ending upgrade path.

 

[Jawed]

Why can't a DAC include the kinds of technology that is apparently seen in the microRendu?

 

[Beolab]

Can someone comment on the pure SQ diffrence btw MicroRendu vs SonicOrbiter SE ?

 

[Beolab]

Read the WhatHifi DAVE review and find this statement to be very funny, because they show their un knowledge about the digital pre amp in the DAVE, that it is the same sound even if you set the volume at -30db as at pass through, it is exactly as transparent no difference in SQ or bit depth loss.

Or what are they purposing by saying;

"The DAVE even makes a decent digital preamp though, as with most such products, a dedicated high quality preamp will work better still. Let’s not forget a suitably talented preamp is likely to cost a similar amount to the DAVE itself."

It is impossible to create a separate analog pre amp to sound more transparent further down the signal path, and this only creates misunderstandings buy 99% who read this.

 

[x RELIC x]

Read the WhatHifi DAVE review and find this statement to be very funny, because they show their un knowledge about the digital pre amp in the DAVE, that it is the same sound even if you set the volume at -30db as at pass through, it is exactly as transparent no difference in SQ or bit depth loss.

Or what are they purposing by saying;

"The DAVE even makes a decent digital preamp though, as with most such products, a dedicated high quality preamp will work better still. Let’s not forget a suitably talented preamp is likely to cost a similar amount to the DAVE itself."

It is impossible to create a separate analog pre amp to sound more transparent further down the signal path, and this only creates misunderstandings buy 99% who read this.

Exactly. The "amp/pre-amp" section in Rob's designs is as transparent to the DAC as you can get with a lot of clean power output, even for the Mojo. I'm reading all over the place that MANY simply don't understand this, especially in most professional reviews of Rob's designs. Adding more components will always add distortions, not take away. :rolleyes:

 

[esimms86]

Can someone comment on the pure SQ diffrence btw MicroRendu vs SonicOrbiter SE ?

The Computer Audiophile review is your best bet for reading about comparisons between the SOSE and the microRendu. Essentially, the microRendu is more analog sounding with better detail and lifelike, room shaking(if you're playing into a room) bass. It's just more...real sounding. Best server I've ever heard. Yes, I own both.

 

[Sonic77]

Chord Dave review on the what hifi site for those that are interested.

Got dinged on the build quality.

 

[Articnoise]

  Leaving tomorrow!
 
But I will be getting hold of a microRendu to try - I am intrigued, particularly as the observation is that it is smoother, which is indicative of lower RF noise. Its a puzzle as battery supply with no earth connection ought to be the lowest way of transmitting RF noise into Dave. But weird things happen with RF, particularly at GHz frequencies. And it still may be nothing to do with Dave, but a symptom of other pick-up in the rest of the system. As I mentioned, YMWV!
 
Rob

 

The battery only isolate from the noise coming from the mains and a laptop is a computer which generate a ****load of its own noise from graphic cards, switching PSU, motherboard, software apps etc. Also there are more ways for the RF to sneak in than thru the mains, as most are airborne. 

 

[Rob Watts]

Airborne RF is not a problem at all for Dave - that's why it is in a solid block of machined aluminium. No RF can get in or out - except via the mains supply and I/O's. The mains supply is extremely well RF filtered, and all IO's are carefully treated for RF.
 
Note that modern homes are a RF hot spot with all the audio turned off, so its essential for good sound quality that treatments are carefully applied everywhere. Note I said carefully - RF treatment on analogue is very different to digital treatments.
 
Rob

 

[TheAttorney]

  Airborne RF is not a problem at all for Dave - that's why it is in a solid block of machined aluminium. No RF can get in or out - except via the mains supply and I/O's.

What about all those holes drilled into the top plate? )
 
Edit: I have a secondary reason for asking about the holes, which will become apparent later. But does it matter if the holes on the left hand side are blocked up by placing anything on top? I.e. are they essential to thermal stability?