CHORD ELECTRONICS DAVE

[lovethatsound]

One of the important factors in communication, is that people have to remember that there are two elements:

1. the message that people want to communicate
2. plus how they communicate that message (this includes the style/pleasant/anger filled/rational/persuasive/factual/etc)

In the past two weeks, the part that lingers in my mind is all the anger, and not the messages that people wanted to communicate.
Ultimately such communication is self defeating, because readers stop reading the posts.

Do you have any documentary evidence that Chord stated that the Blu would be upgradeable?

it was mentioned earlier on in this thread somewhere,plus my dealer told me.☺

 

[romaz]

Hello Roy,

in the meanwhile I have learned that transcoding is different from DoP.
Due to your success with transcoding DSD to PCM files this way sounds interesting for me.
How can I force Roon to apply the transcoding to PCM, which you have already implemented?

I would appreciate your support.

Within Roon, go to Audio Setup and select "Settings" and then "Playback" under MicroRendu.  Under "DSD Playback Strategy", select "Convert to PCM."  Listen for yourself and decide which you prefer.  The advantage of converting DSD to PCM is that you no longer have to toggle between PCM Plus and DSD Plus, which was always a pain to do.  Of course, all stuttering goes away as well.
 
As another experiment, with your regular PCM files, go ahead and switch DAVE from PCM Plus to DSD Plus and you will once again notice that imaging focus degrades but you also get this artificial increased perception of depth and space.  While it is not to my preference, some may find this alternate presentation appealing for certain flat recordings.

 

[romaz]

 
I think the dev speed in DACs has hotted up somewhat. If we slow it down we see other manufacturers catching up / leaving Chord behind. If we speed it up there are those who don't like the idea they bought a DAC that has become bettered by a new version / model 3 months later.
 
The best way to possibly avoid this is:
1. Stick with what you have and enjoy it.
2. Buy a DAC with slot in boards like the C1 or MSB models.
3. Buy and sell on an ongoing basis.
 
Totally empathise with points made here mind.

This is well stated.
 
It is the nature of technology to not stand still and sometimes this advancement can occur at a dizzying pace, faster than some of us are comfortable with.  It is already well chronicled that Rob first conceptualized the significance of 1M taps while at university in the early 80s and he never believed at that time that 1M taps would ever be possible.  During DAVE's development, the Spartan 6 was the best there was and so 164,000 taps, while extraordinary at the time, still fell short.  Whether it caught Rob by surprise or not, I'm not sure, but with the advent of the Xilinx XC7A200T FPGA and its 740 DSP cores, all of a sudden 1M taps became possible and so here we are.  
 
As Rob has stated, going from 500k taps to 1M taps seemed to result in no diminishing return and so obviously, he is curious what pushing further to more taps might result in.  With an array of these FPGAs, he can theoretically push to infinity and beyond.  While 1M taps in theory achieves 16-bit accuracy, because Davina will be capable of 24-bits accuracy, I asked Rob what it would take to achieve this level of accuracy and he told me it would take 250 million taps!  While you could string together an array of 250 FPGAs to achieve this, of course, it would take a massive 2500A PSU and the processing time required would result in a 3.5 minute delay!  I am confident, however, that one day we will see that level of performance but whether it would markedly sound better than 1M taps is anyone's guess.
 
As far as DAVE being obsolete, this proclamation surprises me.  DAVE sounds as stellar today as when I bought it a year ago and even if I had no intentions of ever upgrading to M-scaler, there is still nothing else available at this time that can match its performance.  It remains the truest fidelity DAC that you can buy today bar none and regardless of price and its only true competition are Rob's other DACs.  For people to claim DAVE is obsolete because Hugo2 has closed the gap makes no sense.  While the value proposition of Hugo2 will be unbeatable upon its release, DAVE will still be better and will still have far greater potential.  People have to remember that M-scaler cannot produce sound on its own, it requires a DAC and only DAVE has the capability of fully interfacing with it.
 
I think it's fair that people be able to voice what they would like Chord to come out with and I'm confident they are listening.  I know of no other individual or company that interfaces with their customer base like Rob and Chord have and I believe if there was a genuine ground swell from this community of passionate DAVE owners for Chord to develop something as silly as an M-scaler with a built-in toaster oven so that we could each enjoy freshly baked goods as we listen to music, I have a feeling Chord would do it.  If John felt he could sell M-scalers this way, why not?  As John Franks has stated, Chord is a small company and can only do so much at one time.  If you recall John's concluding sentence from his post 2 days ago, "Have some patience we will not hold back you'll have it all soon enough!"

 

[PhiQuanTu]

Romaz, are you Rob or John in a different nick? You seem to be very knowledgeable about them (even Rob's college dream). A lot of the concerns & questions asked are pretty good points. I wish we could get some answers from the horse's mouth.

 

[Beolab]

@Rob Watts

Just a thaught :

Who will win a "SQ Death-match" between:

1. Hugo 2 or Mojo + M-Scaler ( Davina / Blu 2 )

Vs

DAVE alone in your oppion if you have had time to tested this combination yet?

If you haven't, i am looking forward to your honest answer

The Hugo 2 or Mojo + M-Scaler ( Davina / Blu 2 ) should win if we only are looking on the taprate, but the combo are not toe and toe on the rest specs , so i find this question very interesting.

/

Fredrik

 

[EVOLVIST]

 
 
As Rob has stated, going from 500k taps to 1M taps seemed to result in no diminishing return and so obviously, he is curious what pushing further to more taps might result in.  With an array of these FPGAs, he can theoretically push to infinity and beyond.  While 1M taps in theory achieves 16-bit accuracy, because Davina will be capable of 24-bits accuracy, I asked Rob what it would take to achieve this level of accuracy and he told me it would take 250 million taps!  While you could string together an array of 250 FPGAs to achieve this, of course, it would take a massive 2500A PSU and the processing time required would result in a 3.5 minute delay!  I am confident, however, that one day we will see that level of performance but whether it would markedly sound better than 1M taps is anyone's guess.

 
Will Davina be capable of 24-bit accuracy, though? I know you don't speak for Chord, but on one hand it sounds like you are saying that Davina will be capable of 24-bit accuracy, but only the other, to achieve 250 million taps, it won't be in the Davina chassis.
 
I thought I would ask for you to clarify what you're saying.
 
It doesn't seem like Davina can be too far off. I mean, three-weeks ago Rob said, "Incidentally, I am currently finishing off the first PCB layout for Davina...."
 
What, early 2018 at the latest?
 
With all of the back and forth of the Blu2, I'm reminded of the dream I had in October, of which I wrote:

 

Originally Posted by EVOLVIST /img/forum/go_quote.gif
 
 
I had this dream. No joke, I actually dreamt this, that I went to the Chord factory, and I was hanging out with John Franks and Rob, along with some smoking hot employee of theirs, and for some reason (although this wouldn't happen in real life), Rob let me listen to a "DAVE" that was already at a little over 1million taps, but he explained that he couldn't put it out yet because they need to make money on the 500k, 750k tap models, first, otherwise they would kill their own market. I knew this made business sense - even if it was screwed up - but man, it left me feeling pretty hollow. Rob said, "I mean, why wouldn't I have already reached 1million taps, when all it is is a matter of programing."

 
Interesting. Maybe I've tapped into their taps.

 

[romaz]

Romaz, are you Rob or John in a different nick? You seem to be very knowledgeable about them (even Rob's college dream). A lot of the concerns & questions asked are pretty good points. I wish we could get some answers from the horse's mouth.

Much of what I have posted including the now famous 1M taps threshold that he first conceptualized while at university are based on his public posts.  You just have to read through his posts to verify this.  We have communicated often via PM over the months but I have the same access to Rob as anyone else.  He told me he answers every PM he receives (in due time) which is extraordinary if you think about how busy he is.  Rob was gracious enough to spend several hours with me over CES weekend.  Obviously, I had many questions and he was very kind to answer them.  I can say the same applies to John.  These two are true gentlemen.
 
As for Rob's silence, I'm sure he has his reasons.  Give him his space and I'm sure he will respond in due time.  The last time I communicated with him, he was on his way to Japan.

 

[romaz]

   
Will Davina be capable of 24-bit accuracy, though? I know you don't speak for Chord, but only one hand it sounds like you are saying that Davina will be capable of 24-bit accuracy, but only the other, to achieve 250 million taps, it won't be in the Davina chassis.
 
I thought I would ask for you to clarify what you're saying.
 
It doesn't seem like Davina can be too far off. I mean, three-weeks ago Rob said, "Incidentally, I am currently finishing off the first PCB layout for Davina...."
 
What, early 2018 at the latest?
 
With all of the back and forth of the Blu2, I'm reminded of the dream I had in October, of which I wrote:

 
 
Interesting. Maybe I've tapped into their taps.

There are 24-bit ADCs in existence now (144 dB of dynamic range) and so this is not new although most ADCs have an effective DR of only about 21 bits (or 126dB of DR).  Rob has previously posted Davina will be 24-bits but whether this will be an effective 24-bits, I don't know.  
 
Regardless, most agree that the human ear can only resolve about 23 bits (or 140dB of DR). This is why it is somewhat of a joke for any DAC to boast of extreme dynamic range.  There is a DAC that is famous for claiming a DR of 28.5 effective bits (173dBFS) which is an irrelevant spec even if this measurement is accurate. 
 
On a more practical level, Rob has used "hiss level" to describe the significance of DR and he has documented this several times in the past here on Head-Fi.  The goal is to push your DR to a level that you are no longer able to hear any amount of hiss with even the most sensitive transducer (usually a super-sensitive IEM) and so that is what he did with the DAVE, which has a measured DR of 127dB (A-weighted) or about 21 bits of DR.  He said he could have pushed for a higher number just to pad the spec sheet but there was no point in doing so.  He said that with only the most sensitive set of IEMs can you hear hiss with the DAVE and so with just about any headphone or speaker that most people would use, the hiss level of the DAVE would be at an inaudible level.  Of course, with this DAC that claims a DR of 173dB, since this DAC is incapable of driving headphones or speakers directly and must rely on an outboard amplifier, this wonderful spec becomes completely buried by the DR of the amp, which often is no better than about 100dB (or about 16 bits of resolution).
 
Of course, true analog sound (before it was finitely sampled by the ADC) is of infinite DR but since humans can only hear 23-bits of resolution (or 140dB of DR), then ideally, that should be the theoretical target of any audio system.  Since the dynamic range of most concert halls has been estimated to be between 80-100dB and since the dynamic range of most amplifiers is generally about 100dB, this could serve as a more practical target which equates to about 17 bits (essentially, the DR of a CD).  This is all well and good provided that you can accurately recreate the analog waveform at that resolution to begin with and this is where all DACs fall short but also where only Rob's DACs truly excel.  
 
Where every other DAC strives to translate the compromised digital waveform created by today's compromised ADCs, Rob's DACs, through their very long tap length filters, attempt to recreate the original analog signal before it was compromised by the ADC.  This is where the number of taps matter.  For a 16-bit analog waveform, Rob has calculated you need about 1M taps to recreate this waveform.  If you're going to go for broke and shoot for 24-bits, then you will need 250 million taps.  At what point does the number of taps no longer matter?  That is what Rob hopes to find out and he realizes that it's not just about the number of taps but how they are utilized if musicality is the ultimate goal.  His sole purpose for Davina was never to make a commercial product out of it but to answer some of these questions.  
 
This is from Rob himself and is taken from an interview Rob did for www.the-ear.net.
 
"OK, the interpolation filter’s job in a DAC is to re- create the missing parts of the original analogue signal - the signal in between one sample and the next. This is done with an FIR filter. In a simple way, a FIR filter consists of a data memory (this stores previous data samples) and a coefficient memory (this is a fixed memory with all of the coefficients that the filter algorithm has created). To create an intermediate data, you simply multiply and add all of the stored data samples with a particular coefficient, and once you have added all of the values you end up with the intermediate value you need. Now in the early days, you used a delay line to store the previous data samples, and you tapped into this delay line in order to access the stored data. Hence the word taps.

So why is it only me that goes on about taps and stuff? The problem is about understanding - when I was at university, I studied electronics. But I was passionate about audio, and was interested in the physiology of hearing. I thought if I understood that, I could make better audio electronics. One of the things I was very interested in was how the brain processed the output from the ears. Now we take our hearing for granted, but the brain does some amazing things to give us auditory perception - separating individual sounds into separate entities with placement data (where are sounds located) is an amazing feat, requiring considerable processing. And we know very little as to how the brain does this. Anyway, what I did learn was that transients were a very important perceptual cue, and that the timing of transients was crucial. From transients the ear brain locates sounds in space, it is also used to compute pitch (particularly for bass) and it’s used for getting the timbre of an instrument. I spent a lot of time researching this in the psychology library, which was close to my hall of residence. Anyway, one of the courses in electronics was Whittaker Shannon sampling theory. And this is the basis of digital audio. From this it is a fact that if you had an infinite tap length FIR filter you would perfectly reconstruct the original bandwidth limited analogue signal in the ADC. It would make no difference if it was sampled at 22uS or 22pS you would have the same digital signal. But it was very clear to me that having a limited tap length would create timing errors. And I know from my studies and from my own listening tests that that would be a major subjective problem.

Now, unfortunately, nobody else has recognised this problem for two reasons. One is electronic engineers do not study hearing, and the second problem is that they are stuck on the idea that filters are a frequency domain problem and not a time domain problem. So if you design a filter where your only concern is frequency, then a 100 taps or so is enough. But if you think from the timing perspective, it categorically is not enough. What I have done is to make no assumptions about whether something makes a difference to the sound unless I actually do a listening test. And listening to increasing tap lengths always improves the sound quality. With Dave, I am at 164,000 taps, and I know that that is not the end of it and that further improvements are possible."

 

[analogmusic]

Roy, could you clarify, that 1 million taps is what is needed for 16 bit i.e. CD quality
 
While more taps are needed for 24 bit files, the file size puts me off, and frankly, never heard a huge increase in SQ to justify any spend on 24 bit music files or hardware, once I heard what Hugo was capable of with 16 bit.
 
CD is very good when decoded properly. I expect that Davina encoding 16 bit will be all that is required to enjoy music.
 
yesterday night : listening to "New Jersey" Bon Jovi, the entire album, on Mojo. Really enjoyed this old recording brought to life by Mojo in my car. Wasn't really thinking, "now how would it sound on Dave" as I was simply enjoying the energy of the band and how much fun they had when recording the album, as Rob said, forgetting about the sound, and enjoying the music.

 

[romaz]

Regarding the premature proclamation of DAVE's obsolescence, there is one thing I am hoping will happen with DAVE in the not so distant future that will further enhance its value and appeal for those those of us who own a DAVE but also for those who are looking for a DAC and are wondering if Hugo2 is good enough or whether they should spend more for DAVE.  Rob has stated in the past that DAVE is capable of being upgraded via a code update but that he would never do it unless it resulted in a significant improvement to DAVE.  Perhaps, that time is now for the following reasons:
 
1)  When paired with M-scaler, many of DAVE's DSP cores will now sit idle and I have wondered if these cores can be re-purposed for greater things?  Certainly, it would be ideal for DAVE users who don't plan to upgrade to M-scaler to have Hugo2's improved filters.  Some (Beolab, BMichels, and Jelt2359) have proposed filter options to tailor DAVE"s sound signature (warmer, cooler, neutral, etc) to help balance the tonal deficiencies in one's system.  dCS has such options.  
 
2)  Improving the SPDIF input.  As Rob has stated, SPDIF is not synchronous to DAVE's clock (only USB is) and must go through DPLL first.  Since M-scaler must use the SPDIF inputs, it would probably make some difference in SQ to make the SPDIF inputs as good as USB.
 
3)  Code for better remote functionality.  Rob stated that he ran out of time to properly code for DAVE's remote and so only a few of the remote's buttons actually function.  It would be great if DAVE had a fully functioning remote.
 
I'm sure there would be some nominal cost for such an upgrade which most of us will probably be happy to pay for.

 

[romaz]

  Roy, could you clarify, that 1 million taps is what is needed for 16 bit i.e. CD quality
 
While more taps are needed for 24 bit files, the file size puts me off, and frankly, never heard a huge increase in SQ to justify any spend on 24 bit music files or hardware, once I heard what Hugo was capable of with 16 bit.
 
CD is very good when decoded properly. I expect that Davina encoding 16 bit will be all that is required to enjoy music.
 
yesterday night : listening to "New Jersey" Bon Jovi, the entire album, on Mojo. Really enjoyed this old recording brought to life by Mojo in my car. Wasn't really thinking, "now how would it sound on Dave" as I was simply enjoying the energy of the band and how much fun they had when recording the album, as Rob said, forgetting about the sound, and enjoying the music.

Yes, this is what Rob has said all along.  Ideally, you need an infinite tap length filter but for 16-bits of resolution, he has said 1M taps should get you to a point where you ears can no longer tell that information is missing compared against the original 16-bit waveform.  As I stated, our ears can hear better than 16-bits which is why Rob is pushing to explore what more taps will bring and why Davina will be invaluable for him.  Will 24-bits be necessary?  Here is what he told me some time ago:
 
"Ok it's the coefficient accuracy that we are looking at. This does not mean that the overall accuracy is the same; it may be better depending upon the test signal I guess. Anyway since function halves for doubling the length so 24 bit would be a quarter of a giga tap! But I guess law of diminishing returns will step in. Again that's the benefits of davina; I will have a 768 file properly decimated to 48 then wta scaler back to 768 and I will get an idea of how much losses actually are.
 
Rob"

 

[EVOLVIST]

Yeah, really, the only way I'm digging on a 24/96 file or 24/192, or DSD in its various permutations is because it's a new remaster in that format that you can't get via RBCD. Otherwise, right, for the most part, the differences in sound quality are lost to me. And really, 9 times out of 10 that's what you're getting with 24-bit recordings, from HDTracks, Pono, or whatever, is a new remaster. I'm convinced that most people are going ga-ga over the nice dynamic HDTracks remaster instead of the difference between 16/44.1 and 24/192.

There are really very few of us that have heard what a full 16-bits can give (what Chord is trying to change), so it often makes me think why bother with 24-bit music anyway, unless it's for the aforementioned reasons.

As Roy pointed out, what can the ear hear, and what can it not? Even beyond what the ear takes in, and what the brain encodes, I have doubts that these higher frequencies and extra bits make much of a difference at all.

That's just my take on it. I could be wrong. Just as, when I first got my DAVE I thought crap in/crap out. But then there are some muddy recordings that I have that the DAVE combo of HD800s/Focal Utopia unmuddied some recordings to where they actually did sound better even in just 16/44.

There are many variables as to why that is for some recordings and not for others, including, but not limited to, compressors used in the actual tracking/mixing, the condition of the analogue tape, the amount of digital processing pre and post mix, the EQ of the mastering process, etc.

Is 16-bit enough for playback? I'm pretty sure it is. Does it take an infinite amount of taps or just 1 million? I haven't a clue yet. I haven't hear either at the moment.

 

[Rob Watts]

 

Having spoken with Tom Vaughan, one of Chord's engineers who walked me through Hugo 2's feature set at CES, he described the "Hugo" and "Hugo HF+" filters as being roughly equivalent to "DAVE without HF" and "DAVE with HF", respectively.  Having spoken to Rob about this, the Hugo and Hugo HF+ filters are a further optimized version of DAVE's filters meaning that they are actually better than DAVE's filters.  If I recall correctly, should someone buy an M-scaler for their DAVE, Rob's improved filtering will already be incorporated into M-scaler and is a key reason why the M-scaler sounds better than one would expect from just a 6x increase in TAPS.  In Rob's words, "it's not just the number of TAPS, it's how you use those TAPS."  With M-scaler, there will be no toggle between "HF on" and "HF off" because M-scaler will only sound best one way -- with HF off.

Very interesting. I wished Rob Watts was more active on HeadFi to talk about his achievements with Hugo2's filters in relation to DAVE's. Does anyone have a video of Rob's presentation at the Venetian? Although I need another proper desktop DAC (without a battery and USB micro etc), the Hugo2 narrowly missed my mark and I'm not sure I want to replace my smaller Mojo with the larger Hugo 2.

Yes life has been extremely busy - last week was finishing production code for Blu 2 (tested and delivered to Chord) and finishing Davina PCB (just front panel left to do).
 
I have just arrived in Japan with John, picking up another award for Dave and next week I plan to do a full posting about Hugo 2 on my blog showing all the slides in full plus adding some more comments.
 
But just to clarify - the Hugo 256 FS filter is identical to Dave's - and with the M scaler and Dave the HF filter option is best off as I have innately improved the stop-band rejection - so noise isn't needed to be filtered away because it's already gone. That was why it was odd that 44.1 sounded better overall with the filter on, as my expectation was for no change as rejection of out of band noise was already 130 dB, and in the critical areas (around multiples of the sample rate) with the most noise present, was about 150 dB rejection.
 
Rob

 

[EVOLVIST]

Yes, this is what Rob has said all along.  Ideally, you need an infinite tap length filter but for 16-bits of resolution, he has said 1M taps should get you to a point where you ears can no longer tell that information is missing compared against the original 16-bit waveform.  As I stated, our ears can hear better than 16-bits which is why Rob is pushing to explore what more taps will bring and why Davina will be invaluable for him.  Will 24-bits be necessary?  Here is what he told me some time ago:

[COLOR=0000CD]"Ok it's the coefficient accuracy that we are looking at. This does not mean that the overall accuracy is the same; it may be better depending upon the test signal I guess. Anyway since function halves for doubling the length so 24 bit would be a quarter of a giga tap! But I guess law of diminishing returns will step in. Again that's the benefits of davina; I will have a 768 file properly decimated to 48 then wta scaler back to 768 and I will get an idea of how much losses actually are.[/COLOR]

[COLOR=0000CD]Rob"[/COLOR]

It's a fantastic excerise, and I would love to be proven wrong, but I guess that's the game, right, to see if pushing 24-bits is audible? I would be taking a lot if faith to say that it is. I have read no sound proof that 24-bits gives an audible difference at all. Not even based upon DBs.

 

[romaz]

Yeah, really, the only way I'm digging on a 24/96 file or 24/192, or DSD in its various permutations is because it's a new remaster in that format that you can't get via RBCD. Otherwise, right, for the most part, the differences in sound quality are lost to me. And really, 9 times out of 10 that's what you're getting with 24-bit recordings, from HDTracks, Pono, or whatever, is a new remaster. I'm convinced that most people are going ga-ga over the nice dynamic HDTracks remaster instead of the difference between 16/44.1 and 24/192.

There are really very few of us that have heard what a full 16-bits can give (what Chord is trying to change), so it often makes me think why bother with 24-bit music anyway, unless it's for the aforementioned reasons.

As Roy pointed out, what can the ear hear, and what can it not? Even beyond what the ear takes in, and what the brain encodes, I have doubts that these higher frequencies and extra bits make much of a difference at all.

That's just my take on it. I could be wrong. Just as, when I first got my DAVE I thought crap in/crap out. But then there are some muddy recordings that I have that the DAVE combo of HD800s/Focal Utopia unmuddied some recordings to where they actually did sound better even in just 16/44.

There are many variables as to why that is for some recordings and not for others, including, but not limited to, compressors used in the actual tracking/mixing, the condition of the analogue tape, the amount of digital processing pre and post mix, the EQ of the mastering process, etc.

Is 16-bit enough for playback? I'm pretty sure it is. Does it take an infinite amount of taps or just 1 million? I haven't a clue yet. I haven't hear either at the moment.

You're right.  It is only now with 1M taps that we can hear just how good 16-bits is supposed to sound.  When you get to hear M-scaler, you will hear for yourself just how much is missing with DAVE and how much more is missing with other DACs.  I also agree that the quality of the performance, the recording and the mastering technique are much more important than whether it was recorded in DSD, DXD or 16/44.  I also think that those who go to great lengths to properly record and master a performance today are much more likely to go for higher bit-rate and sampling and so often, people will assume that something sounds great because it was recorded in DSD or DXD.