[sajunky]

At this point, what is the "right technology for music", if at all such a unique option exists, is still inconclusive so I wouldn't jump into conclusions on that part.

Better do not jump in, as you would have to give explanation why R2R technology is not right for a music and specifically what aspects are still inconclusive.
.LOL.

 

[manueljenkin]

Better do not jump in, as you would have to give explanation why R2R technology is not right for a music and specifically what aspects are still inconclusive.
.LOL.

I'm not sure if it's a comprehension problem or your narrative skewing, but nothing in my post mentions R2R in general to be inapt for music. You're the one claiming things should be done in a specific fixed way you assume to be the ultimatum, and I mentioned you there are other ways.

Regarding absolute confidence of perceivable fidelity performance, one would have to analyse every single input signal scenario (non linear performance) and also have to form proper bounds on limits on perceivability, the latter which is still an area of research with not much conclusions.

On the other hand, if you want to push technical performance the problem scenario with the basic types of r2r and sign magnitude r2r were discussed by @dCS James in earlier posts.

 

[sajunky]

I'm not sure if it's a comprehension problem or your narrative skewing

It is a time you can start entertaining others, as in my opinion you have done all.

 

[killaHz]

Perhaps the question is being asked rather rudely, but I think it’s an interesting question: why is DCS going to such lengths to solve distortion problems in a DAC technology that basically nobody uses anymore? Especially since some of what they’re doing at least superficially resembles what D/S DACs are doing.

 

[ThanatosVI]

Perhaps the question is being asked rather rudely, but I think it’s an interesting question: why is DCS going to such lengths to solve distortion problems in a DAC technology that basically nobody uses anymore? Especially since some of what they’re doing at least superficially resembles what D/S DACs are doing.

Well, probably because the end result sounds good.
And I wouldn't say nobody uses it anymore, quite many R2R Implementations are used in modern DACs

 

[dCS James]

Well, it’s been a busy weekend and this week has completely gotten away from me so far, but I’ll attempt to cover some of the points raised:

Firstly, there seems to be a misunderstanding about what we are trying to achieve – we are not trying to “fix” R-2R, or “fix” Delta/Sigma – we are trying to make the best DAC for audio that we can – and due to the high bar we’ve been setting since the early 90’s, there is an element of marginal gains at play – but why shouldn’t we continue striving to enhance performance?

To improve the state of the art you need subject matter expertise. Only by understanding the benefits and limitations of various approaches to A-D and D-A conversion can we design a DAC that we feel is best in class. Of course what we deem as a critical performance requirement may not be the same as other manufacturers or hobbyist engineers, however we are one of a select few manufacturers who have designed world-class ADCs, DACs and sample rate converters for both studio and consumer use.

So, we are trying to point out the issues that exist in common architectures, which to sum up are:

You can have lots of weighted current sources at a low sample rate – the challenge here is matching these current sources, and keeping them matched over temperature variations and time. For audio, the side-effects of this is that any errors in this matching cause unnatural distortion (due to correlation), which the human auditory system is very sensitive to. On the plus side, because we don’t have to run that fast, jitter is less of an issue.

You can have conceptually a single current source, and run it at a much higher sample rate. This fixes the matching issue (because it’s self-referencing – on or off, and any drift will manifest as DC rather than distortion). Unfortunately to deal with the quantisation noise generated, you have to heavily noise shape this and move it up in band. This can cause issues because if you keep the clock at sensible rates, the quantisation noise is very close to the audio band, and if you move it too high in frequency jitter becomes a real issue (due to switching noise), at which point you may have to perform some quite horrible (and sometimes impossible) maths to match rates.

What the Ring DAC does is effectively a hybrid – the clock can run at sensible rates (so 3-6MHz) and the noise shaping can be gentle, but because we have multiple codes to represent, we need a way to match them exactly - whilst bearing in mind that components age, temperature can become a factor and so on. This is the job of the mapper, and it has numerous attractions, including distributing DAC errors away from where we are interested (audio frequencies) to where we are insensitive (very high frequencies), without altering the data presented to it, whilst at the same time ensuring all the sensitive components age in the same way.

We can definitively state the that the Ring technology is not multiple DSD streams, and is not random – if you read the writeup, we even say “may appear random”.

It is quite correct to say that you cannot decorrelate noise that is already part of the signal. However, it is worth thinking about this as a philosophical point. One view is that noise shaping and filtering are evil because they somehow ‘guess’ and don’t reproduce the ‘original’ signal, and the ‘fewest steps must be the best’. Now, this may sound strange coming from us, but one of our beliefs can be attributed to Einstein – “Everything should be made as simple as possible, but no simpler”. So what is the original signal we are trying to reproduce?

We’ll cover this more in the filtering article, but here is a genuine question – on the scope shots supplied, we can see some effects that are clearly the lack of sensible filtering – if this is meant to show averaging, then the steps are way, way too big (consider all the discussions on matching have been < 1%). Does anybody think that the stair steps seen in the photographs are beneficial in any way? If so, how? Where do they come from?

 

[sajunky]

You are not trying to fix R2R DAC, Ring-DAC has nothing to do with PCM direct decoding. You are trying to fix Delta-Sigma DAC. And this is a problem as chip DS DAC's now are doing similar things and cost much less.

 

[ThanatosVI]

You are not trying to fix R2R DAC, Ring-DAC has nothing to do with PCM direct decoding. You are trying to fix Delta-Sigma DAC. And this is a problem as chip DS DAC's now are doing similar things and cost much less.

And sound much worse..

As they said they don't try to "fix R2R or fix D/S" but try to create the best sounding DAC they can. And evidently they were very good at it so far.

If you are looking for the most cost efficient product, then yeah go with a DS solution, these are quite affordable nowadays as you said.

 

[Pondoro]

I work in a technical field. Egos are rightly large, as the practitioners are very smart and generally well educated. People argue. But I rarely hear a specific technology get attacked. Technologies rise and fall with the era's technical abilities. Example - organic fiber composites. No one would have put them in an engine in 1962. (Fan belts were an exception, some had steel, others nylon, fibers.) Organic fiber composites are all over engines now. I am sure people had many arguments, component by component, over what was better, metal or composite? But I am sure no one ever made a blanket statement, "organic fiber composites are out of date!" People argue the merits. Yet organic fiber components had existed a long time - adobe! Would the fact that adobe is a bad material to use in an automotive engine cause someone to say, "organic fiber components are no good?" I doubt it. So a technology, like any of the DAC architectures mentioned here, may be the best in 1987, then not the best, then reappear years later. The author is trying to explain the barriers of all technologies and how a designer must mitigate those barriers. Telling us that "this technology is no good" is very short sighted. There are three questions that get confused in the arguing: 1) Is the explanation technically correct? 2) Does the resulting DAC that is built by this author sound good? 3) Is it the best for me at it's price? Statements that R2R or DSD or XYZ is bad, or good, or cannot be successfully implemented show a lack of humility.

 

[sajunky]

We’ll cover this more in the filtering article, but here is a genuine question – on the scope shots supplied, we can see some effects that are clearly the lack of sensible filtering – if this is meant to show averaging, then the steps are way, way too big (consider all the discussions on matching have been < 1%). Does anybody think that the stair steps seen in the photographs are beneficial in any way? If so, how? Where do they come from?

Yes, these steps are way to big, it is something to simple to be a simple model as Einstein said.

In an ideal situation a ladder calibration table should be created at a factory and burned in NVM. If a first 6 MSB's are calibrated, it leads to the six mathematical operations against a corrected value in a table for each sample. It is probably to complex job for FPGA. A quite easy way is to randomise errors instead. It is difficult to interpret a screenshot, as a picture is built from multiple traces. It is clear however that randomisation process is also simplified. It looks like that a six MSB's are randomised in the way that for each 6-bit MSB combination a six lower significant bits take not the same value on every sample, but there are scramled according to some internal function. It create a specific stepping pattern with random steps inside each step. Not really a clean solution. It measure well, as dashboard brings an averaged result, but there are users reporting that they hear a strange noise running parallel to a music, like listening to sea weaves on the beach. In summary, Holo Audio is not on my purchase list.

 

[sajunky]

If you are looking for the most cost efficient product, then yeah go with a DS solution, these are quite affordable nowadays as you said.

Ring DAC is a DS solution. It can sound good in a DS way.

 

[ThanatosVI]

Ring DAC is a DS solution. It can sound good in a DS way.

It was quite obvious from the comment that I was talking about a DS Chip solution...

 

[manueljenkin]

Considering how deviated from a basic delta sigma modulator, both the DCS and the many of the other chip dacs (like ess) are, I am not sure if we could call them as delta sigma either.

 

[sajunky]

Considering how deviated from a basic delta sigma modulator, both the DCS and the many of the other chip dacs (like ess) are, I am not sure if we could call them as delta sigma either.

Delta-Sigma modulator remains unchanged, it is the place where sound is simplified.

 

[mammal]

For those interested in a detailed review of dCS Bartók driving AB-1266 and how it compares to Chord Hugo TT2 (with M Scaler) have a look here.