CHORD ELECTRONICS DAVE

[106099]

Is there some Microrendu review by a magazine like HiFi News somewhere?
 
Paul

 

[shuttlepod]

  Is there some Microrendu review by a magazine like HiFi News somewhere?
 
Paul

Paul, I don't think you are going to find any "traditional" review of the mR, with the exception of the partial review (Part 1) on computeraudiophile by Chris Connaker. Connaker promises a Part 2 fairly soon, once he's done with the Munich show. But there are a number of threads on the forums at computeraudiophile where users are posting impressions. Realize that this product only began to ship in the last ten days. I am personally going to wait a number of months for things to shake out before making any decision.

 

[romaz]

 
I hadn't realised how much the aspect of timing fidelity is being emphasised, in the MQA promotional copy.
 
Since timing accuracy also happens to be a major emphasis of Rob's WTA filter approach, I am curious as to whether or not the two approaches are compatible, and I am also curious how MQA might be 'capturing' (digitising) timing details with (what seems to be being implied) greater accuracy than all other existing ADC approaches.

Improving timing fidelity has always been the promise of MQA so they are now not saying something they haven't been saying all along.  What they have always purported to address were the timing issues that were the consequence of ringing artifacts introduced by ADCs.  An MQA file is supposed to have embedded metadata that will inform an MQA decoder-equipped DAC what ADC was used in the recording.  That DAC is then supposed to apply a fix to compensate for the deficiencies of the ADC.  How does this occur?  Through oversampling, a word we've seen used a lot lately.
 
Unlike HQ Player, however, oversampling occurs only in the PCM domain and is a PCM-only format.  MQA is supposed to oversample to the capability of the DAC being used.  If it's an iPhone-type device with limited oversampling capability, improvement will be heard but only to a smaller degree.  If you have a DAC capable of higher sampling, like 768 kHz, then MQA is supposed to take advantage of that capability with even better results. R2R DACs, because of their limited ability to oversample will not be an ideal candidate for MQA.  Once again, what does better timing fidelity get you?  Depth and air and scale -- the same things that both HQ Player and the DAVE claim to improve upon.  The world has tired of flat 2-D sound.
 
Although these 3 approaches (HQ Player, MQA, DAVE) strive to achieve the same benefit (timing fidelity) through oversampling, you will have to decide which path you will take because from my understanding, each approach is philosophically and fundamentally different and not complementary.  I have no expectation that Chord will be implementing MQA decoders in any of Rob's DACs and Rob himself has said he has no plans to do so.  Having heard an MQA demonstration with a Mytek Brooklyn DAC, which incorporates an MQA hardware decoder and is capable of 384 kHz oversampling, I was able to toggle the MQA decoder on and off and so it was easy to A/B the impact of MQA decoding and while there was a noticeable improvement in depth and air, I would say that at no time did I feel I was listening to anything as good as what I get with the DAVE.  If I am to provide my own interpretation of the situation, to ask Rob to alter the DAVE and shift away from what the DAVE does now through oversampling and instead take the MQA route would be to take a step backward.  You could argue the same thing with HQ Player.
 
Because MQA-files are PCM files at their core, they will play back on any PCM-capable DAC including the DAVE and so there is no worry about playback compatibility.  There are MQA files you can download and listen to on the 2L test bench if anyone is interested.  As a wrapper, MQA files can be useful for DAVE owners as their compression/decompression scheme allows for the streaming of hi-res files through sites like Tidal without the need for gobs of bandwidth.   In the demo I heard at Las Vegas, using a beta version of Tidal, a 24/384 MQA file was streamed and decoded from a server in Europe with no drops or skips encountered.  
 
Bob Stuart and company have worked hard to disseminate MQA and establish it as a standard and at CES in January, there certainly was a lot of MQA buzz.  If you are to believe Robert Harley at TAS, then you are probably thinking MQA is the best thing since sliced bread and as a standard, it is a done deal.  While Warner has signed on and more are likely to follow, it's interesting to note that so far, only 2 companies (Meridian and Mytek) are shipping MQA-enabled DACs.  Where are the rest?  As has already been stated, you are not likely to see R2R DACs touting MQA capability and the HQ Player camp probably sees no need for MQA either since they would prefer to upsample to DSD.  It was once believed that MQA decoding could take place at the music server level just like HQ Player and indeed, Auralic had announced MQA capability through a firmware upgrade for their Aries back in January.  At the 11th hour, however, they removed this feature reportedly at the request of MQA.  The rumor I heard was MQA had decided proper decoding needs to take place at the DAC and not the music server and that the DAC must have a hardware MQA decoder for this to happen.  If this is true, then MQA is not a done deal since this retrofit is probably not an easy thing to do.  Moreover, DAVINA only complicates this further as it will compete with MQA.  Why try to compensate for a technically inferior ADC when you can just build an ADC without these deficiencies?  In the same way that a bunch of movie studios signed on to support betamax when it first came out, as we know, it was VHS that eventually prevailed.

 

[rkt31]

what happened to the download link of dacs comparison ? right now i have only the recordings  made of dave's analog output as a means of experiencing the high end dacs . in India it is almost impossible to audition such an expensive gear.

 

[DaveRedRef-III]

.......Even though it was agreed by all that Betamax was the superior protocol.

It seems to me that MQA is initially considered by audiophiles as a competitor to their Dac product. Particularly, it seems, for owners of Dave. It is unfortunate because it shouldn't be like that. MQA is less about resolution than it is about timing correction in my view. If MQA gives an MQA capable Dave a timing corrected master it will sound better. At least that is how I understand the process.

Timing problems have plagued studios ever since a producer started looping effects units and compression units into the mix. It can get really bad in digital especially when there is limited processing power for a complicated mix. Even with good power headroom modern DAWs have timing compensation built in. This was not always the case and has only become mainstream in the last 15-20 years. This is not a problem for classical music as the recording is simple and real world but contemporary recordings are the product of all sorts of weird and wonderful setups which include processing which may not be stereo universal in each case. Now I don't think for one moment that MQA,s mastering process can fix all ills that occur in the long recording process but I am willing to buy that they have identified one or two aspects of the mastering process that are continually misaligned in the finished digital product.

 

[DaveRedRef-III]

To continue

I see Daves superior resolution as being beneficial to solving the industry limited process of d/a conversion in terms of resolution. However Dave will not correct timing errors that are frozen in the master copy. It will only serve to 'highlight' that misalignment. That is where I think MQA is coming from.

 

[JaZZ]

Let's not forget that DAVE's «reconstruction» algorithm is made to reconstruct the original signal before analogue/digital conversion. So, as I see it, it shouldn't be as sensitive to ADC imperfections as conventional DACs – hence wouldn't benefit as much from MQA.

 

[DaveRedRef-III]

It would be interesting to be able to test that Jazz and I hope we get the chance one day. My listening experience with Dave so far has been enhanced to the point that I am 'more' aware of inaccuracies and weaknesses in the recording process but the resolution is so good that when the feel of the performance is hot Dave shows how superior and natural the performance was. I am far more aware of intonation errors in vocals. I can also discern that multi sampling 44.1 before it enters Dave is not as natural as keeping it at 44.1 which is the opposite of my experience before Dave. I am also more acutely aware of sloppy musicianship, in particular lack of left - right synchronicity between musicians on occasion but maybe that is because a sloppy producer ran a reverb bleed on one instrument and a multi processor chain on another instrument and panned them left and right (in other words it is timing errors in the production process accentuating a less sloppy performance).

I guess my point is that Dave is so transparent I wonder if the benefits of MQA might be 'more' apparent.

 

[JaZZ]

Dave...
 
...of course DAVE is a highly revealing DAC, far from forgiving in terms of recording flaws. What I meant was that ADC imperfections in them would possibly be less of an issue due to the sophisticated filter algorithm based on the 164,000 taps, which tries to reconstruct the original analogue waveform rather than stricty relying on the digital data at hand, as with classic even-numbered oversampling. Rob has once explained that he wasn't trying to preserve the original samples exactly for this reason.

 

[DaveRedRef-III]

I appreciate your point Jazz and I would not be surprised at all if, due to Daves exceptional resolution, it were able present MQA files better than Meridians own flagship product.

 

[Crgreen]

I really don't know much about the technical details, but the impression I get is that. If you want an MVQ enabled DAC, it's not a retrofit or upgrade, you have to but a new DAC with the chip that will provide MVQ facilities. This might be a reason to hold fire if you're thinking of buying a new DAC. I've plumped for the DAVE (expected at the end of the month) because I was so impressed with the Hugo, but I'd assumed that it would rule out MVQ in respect of what I plan to be my last DAC.

The other issue which struck me about MVQ is that although I can see the benefits for something like Tidal and other streaming services, I for one don't plan to start repurchasing discs or downloads of existing recordings just to have them in MVQ form. And I'm sure I'm not alone. I just don't see where a format war will get us.

 

[Crgreen]

Apologies, MQA - I go a bit dyslexic when it comes to all theses abbreviations and acronyms. My eyes just glaze over.

 

[Mython]

 
I hadn't realised how much the aspect of timing fidelity is being emphasised, in the MQA promotional copy.
 
Since timing accuracy also happens to be a major emphasis of Rob's WTA filter approach, I am curious as to whether or not the two approaches are compatible, and I am also curious how MQA might be 'capturing' (digitising) timing details with (what seems to be being implied) greater accuracy than all other existing ADC approaches.

 
 

  I'm neither pro nor anti MQA, in regard to whether or not Chord 'should' or 'shouldn't' implement it, or whether or not it would be competitively prudent to - I'm just curious about the technicalities underlying MQA.
 
I understand that they are accessing original master tapes (didn't Neil Young say he was going to do that, before his store just started selling all the same Hi-Res files as every other Hi-Res download store on planet earth?

)
 
Joking aside, I understand that MQA intend to dynamically 'calibrate' the playback chain (including the digital transport & DAC, unless I've misunderstood) to most accurately mimic the sound of each original analogue master tape (presumably via some kind of metadata embedded within each MQA file?). That'd be all well & good.
 
What I don't understand, however, is how the MQA approach can improve timing accuracy, unless their actual ADC itself (as a seperate entity from any subsequent codec-engineering, further along the process) is somehow a step ahead of all existing ADCs (and, as DAVE fans know, Rob happens to be working on an ADC following some of his WTA principles implemented in DAVE, Hugo and Mojo). The cleverest codec in the world cannot make a silk purse out of a sow's ear, WRT to timing accuracy, if the ADC itself is no better than any other ADC. Perhaps they are merely saying that the MQA codec retains maximum timing information available in the raw ADC'd PCM file.

 
 

Spoiler: Warning: Spoiler!

  Improving timing fidelity has always been the promise of MQA so they are now not saying something they haven't been saying all along.  What they have always purported to address were the timing issues that were the consequence of ringing artifacts introduced by ADCs.  An MQA file is supposed to have embedded metadata that will inform an MQA decoder-equipped DAC what ADC was used in the recording.  That DAC is then supposed to apply a fix to compensate for the deficiencies of the ADC.  How does this occur?  Through oversampling, a word we've seen used a lot lately.
 
Unlike HQ Player, however, oversampling occurs only in the PCM domain and is a PCM-only format.  MQA is supposed to oversample to the capability of the DAC being used.  If it's an iPhone-type device with limited oversampling capability, improvement will be heard but only to a smaller degree.  If you have a DAC capable of higher sampling, like 768 kHz, then MQA is supposed to take advantage of that capability with even better results. R2R DACs, because of their limited ability to oversample will not be an ideal candidate for MQA.  Once again, what does better timing fidelity get you?  Depth and air and scale -- the same things that both HQ Player and the DAVE claim to improve upon.  The world has tired of flat 2-D sound.
 
Although these 3 approaches (HQ Player, MQA, DAVE) strive to achieve the same benefit (timing fidelity) through oversampling, you will have to decide which path you will take because from my understanding, each approach is philosophically and fundamentally different and not complementary.  I have no expectation that Chord will be implementing MQA decoders in any of Rob's DACs and Rob himself has said he has no plans to do so.  Having heard an MQA demonstration with a Mytek Brooklyn DAC, which incorporates an MQA hardware decoder and is capable of 384 kHz oversampling, I was able to toggle the MQA decoder on and off and so it was easy to A/B the impact of MQA decoding and while there was a noticeable improvement in depth and air, I would say that at no time did I feel I was listening to anything as good as what I get with the DAVE.  If I am to provide my own interpretation of the situation, to ask Rob to alter the DAVE and shift away from what the DAVE does now through oversampling and instead take the MQA route would be to take a step backward.  You could argue the same thing with HQ Player.
 
Because MQA-files are PCM files at their core, they will play back on any PCM-capable DAC including the DAVE and so there is no worry about playback compatibility.  There are MQA files you can download and listen to on the 2L test bench if anyone is interested.  As a wrapper, MQA files can be useful for DAVE owners as their compression/decompression scheme allows for the streaming of hi-res files through sites like Tidal without the need for gobs of bandwidth.   In the demo I heard at Las Vegas, using a beta version of Tidal, a 24/384 MQA file was streamed and decoded from a server in Europe with no drops or skips encountered.  
 
Bob Stuart and company have worked hard to disseminate MQA and establish it as a standard and at CES in January, there certainly was a lot of MQA buzz.  If you are to believe Robert Harley at TAS, then you are probably thinking MQA is the best thing since sliced bread and as a standard, it is a done deal.  While Warner has signed on and more are likely to follow, it's interesting to note that so far, only 2 companies (Meridian and Mytek) are shipping MQA-enabled DACs.  Where are the rest?  As has already been stated, you are not likely to see R2R DACs touting MQA capability and the HQ Player camp probably sees no need for MQA either since they would prefer to upsample to DSD.  It was once believed that MQA decoding could take place at the music server level just like HQ Player and indeed, Auralic had announced MQA capability through a firmware upgrade for their Aries back in January.  At the 11th hour, however, they removed this feature reportedly at the request of MQA.  The rumor I heard was MQA had decided proper decoding needs to take place at the DAC and not the music server and that the DAC must have a hardware MQA decoder for this to happen.  If this is true, then MQA is not a done deal since this retrofit is probably not an easy thing to do.  Moreover, DAVINA only complicates this further as it will compete with MQA.  Why try to compensate for a technically inferior ADC when you can just build an ADC without these deficiencies?  In the same way that a bunch of movie studios signed on to support betamax when it first came out, as we know, it was VHS that eventually prevailed.

 

 
 
Quote:

   Why try to compensate for a technically inferior ADC when you can just build an ADC without these deficiencies?

 
 
Thanks for elaborating on my musings

 
 
I agree that that is like 'trying to make a silk purse out of a sow's ear' (old English saying).
 
 

To continue

I see Daves superior resolution as being beneficial to solving the industry limited process of d/a conversion in terms of resolution. However Dave will not correct timing errors that are frozen in the master copy. It will only serve to 'highlight' that misalignment. That is where I think MQA is coming from.

 
 
Mmmm.... but Davina will.
 
 
 
 
 
 
I am far more interested in seeing what difference (I anticipate MASSIVE) Rob's ADC (AKA 'Davina') will make to the eventual reproduction of a recorded performance, than I am in MQA 'compensating' (however cleverly it might do so) for sub-par ADC timing-resolution when transfering from an analogue master tape. That is not to say that I don't think MQA may have a valid, respectable, and useful contribution, in other regards, outside of a Chord digital ecosystem.
 
My guess is that Davina + DAVE will perform an order of magnitude better than even the sum of each of their already-capable contributions.
 
Exciting times ahead, and Rob, thankyou again for all you're doing to advance the state of the recording-&-playback art.
 
 
 
.

 

[romaz]

.......Even though it was agreed by all that Betamax was the superior protocol.
 

Yes, this is true.  It goes to show that what is best and what the market accepts are not always the same and so we will need to be careful what we wish for.
 

It seems to me that MQA is initially considered by audiophiles as a competitor to their Dac product. Particularly, it seems, for owners of Dave. It is unfortunate because it shouldn't be like that. MQA is less about resolution than it is about timing correction in my view. If MQA gives an MQA capable Dave a timing corrected master it will sound better. At least that is how I understand the process.

Timing is a form of resolution.  It is my view that "time resolution" is the more difficult type to master and it is what separates the great DACs from the average DACs.  
 
I believe you've missed the point of my rather long discourse because while both MQA and the DAVE both incorporate their own proprietary technologies to improve timing fidelity, the two technologies are philosophically and fundamentally different and both cannot occur at the same time with the DAVE.  MQA will choose to use the DAVE to oversample to address ringing artifacts introduced by the ADC.  Rob believes this to be irrelevant and he oversamples for completely different reasons.  Here is what Rob had to say specifically about why he oversamples compared to why MQA oversamples:
 
"Its done for a number of reasons - to reduce the timing of transients uncertainty problem, to enable the noise shapers to work at 104 MHz so that the noise shapers can reproduce depth correctly, and finally to allow no measurable noise floor modulation.
 
So there are a number of reasons why I oversample to such a high rate.
 
ADC ringing artifacts is not one of them, as that is irrelevant."
 
Also, JaZZ is absolutely correct.  The DAVE attempts to reconstruct the original analog signal before A/D conversion which is why I think Rob believes that what MQA is trying to accomplish is "irrelevant."  Here is what Rob had to say on the Mojo thread:
 
Except the job of a DAC is NOT to reproduce the sampled data perfectly but to reproduce the original bandwidth limited analogue signal that was in the ADC before the signal was sampled. And to do this one must convert from a sampled signal and convert it to a continuous waveform - and that actually implies infinite oversampling, something that a R2R DAC can't do as they are limited to 16FS oversampling due to speed and glitch problems. That's one reason (there are many others too) why Mojo filters to 2048FS and has its DAC run at 104 MHz, unlike any other non Chord DAC's.
 
Here's more:
 
For a interpolation, or more accurately reconstruction filter, if you use an infinite tap length sinc FIR filter, you will reconstruct the original analogue signal in the ADC completely perfectly with no change whatsoever. But such a filter will have an inifinite amount of post and pre ringing - and this contradiction is solved when you realise that an impulse is not a legal signal, as it contains the same energy at FS/2 as at 1 kHz - so it is not bandwidth limited which requires zero output at FS/2. The idea that pre ringing is audibly bad is mistaken as an illegal (from sampling theory) test signal is being used. The filters that have the maximum pre-ringing using a non bandwidth limited signal is actually much more accurate in that when using a proper bandwidth limited signal will give the least differenct to the original un-sampled signal.
 
So yes you are correct - I am trying to perfectly reproduce the original un-sampled analogue signal, and the best filter is one that has an infinite amount of pre-ringing using an illegal non bandwidth limited impulse response. The WTA filter is much closer to an ideal sinc function, so has huge levels of pre-ringing - but using a proper bandwidth limited signal will return a signal that is closer to being unchanged than any other reconstruction filter available today.
 
And that's the primary reason why Mojo sounds like "you are there" because it more accurately reconstructs the analogue signal before it was sampled.
 
Rob
 
As to the specific problems he sees with current ADCs and why the DAVINA project exists, timing fidelity isn't the only problem and so it would appear that MQA only attempts to be a partial solution.  Here is what Rob had to say about the issues with current ADCs:
 
I have looked at the chips used for ADC's and the best ones offer -122 dB SNR, and THD and noise of -110dB. I am actually targeting 135 dB SNR, and THD and noise to be the same as Dave.
 
But by far the biggest problems are:
 
1. Depth perception - the work on Dave indicates that you need 350 dB capable noise shaping to accurately preserve depth - and current devices are only about -140 dB
 
2. Noise floor modulation - a chip based ADC has a lot of noise floor modulation, and I have solved this problem by using Pulse Array and discrete integrators.
 
3. Aliasing. The conventional view is that aliasing products in the 20-22.05 kHz range is inaudible. But it is not, as this aliasing errors degrade the timing of transients - which are audible.
 
I hope Davina will be less than Dave in terms of price. But since it is a technology prove of concept, its not about cost.
 
Rob
 
The bottom line is you will need to decide which ideology to subscribe to but with Rob's DACs, you can't have both.

 

[romaz]

I really don't know much about the technical details, but the impression I get is that. If you want an MVQ enabled DAC, it's not a retrofit or upgrade, you have to but a new DAC with the chip that will provide MVQ facilities. This might be a reason to hold fire if you're thinking of buying a new DAC.

Some DACs can be retrofitted.  This is what Michal Jurewicz told me he did with the Mytek Manhattan and this DAC will become MQA-enabled next month.  It's true, however, that you can't do it through a software/firmware upgrade because the decoder is a physical piece.