Chords New M -Scaler

[BigAlMc]

10 minutes ago, ElviaCaprice said:

If your going to spend 4k pounds for an m-scaler, then you might as well get a Qutest or Hugo 2 to take full advantage of it.  Otherwise TT2 or DAVE are the other choices.

 

I was very impressed with the QuTest at a hifi show earlier this year. For the money it sounded very natural and very nice. But I've too much love for PS Audio and my Directstream to even consider such, er, infidelity ? at this stage.

 

Anyways risking going off-topic here. Was just hoping for some insight into the relative synergies of super upscaling vs super clocking. 

 

Cheers, 

Alan

[rayl1234]

27 minutes ago, BigAlMc said:

 

Dude I thought you and I were in the same ballpark if not page when we posted at pretty much the same time referring to USB. Guess not! 

 

The Chord defo has the prettier box. It's very cute. 

 

Question stands if anyone wishes to opine with more than sarcasm. 

 

Cheers, 

Alan

 

 

As a former DirectStream person who converted to Chord Blu2/DAVE last fall, I unfortunately agree.... (and matched my actions to that conclusion)

[ecwl]

3 hours ago, adamdea said:

Apologies if this has been covered already, but it amuses me that the scaler thing does not have usb out. Funny how we were supposed to this that s/pdif and its inherent jitter /long term clock matching problem was supposed to be THE issue. And that was at 44.1kHz not 768kHz.

Rob Watts addresses this recently on Head Fi forum. The issues was that he has not had time to thorough test a USB transmitter to ensure it works perfectly to. He thinks to do that adequately to his satisfaction (prototyping, ensuring lack of excess noise, ground plane issues, etc), he would need an extra 6 months. That’s why the HMS has no USB out. His ADC Davina project will have a USB out but who knows when that’ll be ready. 

 

From a jitter, clock perspective, since HMS is meant to be paired with Chord’s newest DACs which are extremely jitter immune, that’s why Rob Watts is okay with not having USB out and sticking with dual S/PDIF. 

[Jud]

35 minutes ago, ecwl said:

From a jitter, clock perspective, since HMS is meant to be paired with Chord’s newest DACs which are extremely jitter immune, that’s why Rob Watts is okay with not having USB out and sticking with dual S/PDIF. 

 

"Extremely jitter immune" is words.  What @adamdea is saying is that technically S/PDIF can't have jitter as low as an async USB interface.  The reason is that the clock is local to the DAC with async USB, while with S/PDIF the DAC clock always has to try to synchronize itself with another clock.

 

This difference of course may not be remotely audible, but that's not what was being said.

[Whitigir]

I still don’t understand the point of M-Scaler.  Why wouldn’t PC be able to do the same, or why wouldn’t an offline conversion do the job ?  I have seen Rob and some one else keep claiming that his filter and algorithms are perfected and superiors to others.  Very interesting 

 

are there other similar filters ? Or is his really superior

[The Computer Audiophile]

19 minutes ago, Whitigir said:

I still don’t understand the point of M-Scaler.  Why wouldn’t PC be able to do the same, or why wouldn’t an offline conversion do the job ?  I have seen Rob and some one else keep claiming that his filter and algorithms are perfected and superiors to others.  Very interesting 

 

are there other similar filters ? Or is his really superior

 

A computer can do it, but in a far less elegant package and requires the user to do far more than connect a couple cables. 

 

Resale value will will probably be higher for the M Scaler as well :~)

[Whitigir]

I meant if you do an offline conversion then it is all algorithms isn’t it ? Timing errors is negated out

[The Computer Audiophile]

Just now, Whitigir said:

I meant if you do an offline conversion then it is all algorithms isn’t it ? Timing errors is negated out

 

Is there an app that does offline conversion using the same filters?

[Whitigir]

Not yet,but I can’t imagine that someone else can’t make such an app

[ecwl]

34 minutes ago, The Computer Audiophile said:

 

Is there an app that does offline conversion using the same filters?

I don't think there is because I don't think the software people think extremely long tap filters that closely resembles ideal sinc filters offer sonic benefits, as evidenced in this forum's discussion so far. They think there are other ways to filter more efficiently computationally that sounds better.

[ecwl]

4 hours ago, Jud said:

 

"Extremely jitter immune" is words.  What @adamdea is saying is that technically S/PDIF can't have jitter as low as an async USB interface.  The reason is that the clock is local to the DAC with async USB, while with S/PDIF the DAC clock always has to try to synchronize itself with another clock.

So this is getting a little more off topic because it touches more on DAC design than Chord Hugo M-Scaler. Rob Watts, the Chord DAC designer thinks that async USB or digital PLL solutions are already excellent for jitter. He thinks most DAC jitter actually comes from the switching noise of the R2R or DSD/PWM or multibit SDM DAC chips. He thinks noise feeds back to the clock and the whole DAC. He also thinks the bigger issues is that the switching noise is signal dependent. This is one of the main reasons he says he designed the pulse array DAC which is like a discrete elements version of ultra-high frequency multiunit SDM DAC using thermometer code with an extra twist which is constant switching, not in the usual sense of dynamic element matching (DEM) but in the sense that the same number of elements are constantly switching so that the switching noise is always the same which would control jitter and also eliminate noise floor modulation.

[jabbr]

1 hour ago, ecwl said:

He thinks most DAC jitter actually comes from the switching noise of the R2R or DSD/PWM or multibit SDM DAC chips. 

 

There is little question that “jitter” or phase noise is a strong function of the IC chip and particular logic family. Lots of focus on getting a clock with perfect specs ...

[rickca]

36 minutes ago, jabbr said:

 

There is little question that “jitter” or phase noise is a strong function of the IC chip and particular logic family. Lots of focus on getting a clock with perfect specs ...

You probably should say this a bunch of times so it sinks in.

[barrows]

1 hour ago, jabbr said:

 

There is little question that “jitter” or phase noise is a strong function of the IC chip and particular logic family. Lots of focus on getting a clock with perfect specs ...

I do not doubt the validity of the above point, but would also suggest, that given a specific "IC chip" and "particular logic family", IE what one is already dealing with,  a way to improve performance can be applying batter a better masterclock (given that the details are attended to).

In no way I am I suggesting avoiding the details of the former in favor of the latter.

In my experience of improving the masterclock (and I do not refer to applying an external clock here, but replacing the existing masterclock(s) with one of lower phase noise), I have yet to do so and not experience an improvement in sonics as a result.

 

I still have not got my hands on any of the really impressive XOs yet though (ones significantly better than NDK 2520 SDA series or Crystek 575 and 957 series, all of which are pretty closely grouped in general for close in phase noise).

 

 

 

[adamdea]

7 hours ago, ecwl said:

So this is getting a little more off topic because it touches more on DAC design than Chord Hugo M-Scaler. Rob Watts, the Chord DAC designer thinks that async USB or digital PLL solutions are already excellent for jitter. He thinks most DAC jitter actually comes from the switching noise of the R2R or DSD/PWM or multibit SDM DAC chips. 

....

but in the sense that the same number of elements are constantly switching so that the switching noise is always the same which would control jitter and also eliminate noise floor modulation.

[as for the first thing -he may well be right, albeit selectively] You mean he thinks that the problme is the thermometer dac elements switching on and off. Not sure I follow- that simply is the output signal.

Anyway I'm genuinely intrigued as to what this means.  How can he have constant switching - if you go from min to max you have to switch on all the elements. So how does he ensure the same number of elements are changing irrespective of the signal- does he have all the elements change but have some of them out of circuit?

Presumably this could be demonstrated on a simple  J est - or do we need to refine the J test to involve maximum changes of dac element values (J test was perfected for 16 bit numbers)

[mansr]

1 hour ago, adamdea said:

Anyway I'm genuinely intrigued as to what this means.  How can he have constant switching - if you go from min to max you have to switch on all the elements. So how does he ensure the same number of elements are changing irrespective of the signal- does he have all the elements change but have some of them out of circuit?

I was thinking the same thing. Pulse array seems to be a term invented by Rob Watts, so it's impossible to find out what it actually is. Typical.

[Whitigir]

Isn’t pulse array similar to S-Master by Sony ? But how does any kind of communication is said to immune to jitter ? That is crazily unrealistic 

[jabbr]

12 hours ago, ecwl said:

This is one of the main reasons he says he designed the pulse array DAC which is like a discrete elements version of ultra-high frequency multiunit SDM DAC using thermometer code with an extra twist which is constant switching, not in the usual sense of dynamic element matching (DEM) but in the sense that the same number of elements are constantly switching so that the switching noise is always the same which would control jitter and also eliminate noise floor modulation

An excersize for the mathematicians among us would be to design a thermometer code which has a constant number of “1”s despite the scalar value.

[ecwl]

24 minutes ago, jabbr said:

An excersize for the mathematicians among us would be to design a thermometer code which has a constant number of “1”s despite the scalar value.

All of you are right. I don't know how Pulse Array DAC works. And Rob Watts only "vaguely" describes it on the Head-Fi forum. But of course, why should he have to share it if most other DAC designers don't share every exquisite detail? Everybody has their secret sauce whether it's for real or not.

 

Anyway, this is my speculation of how it works. But I'm pretty sure I'm wrong in some ways. I thought most of you who have been commenting have read this already since it's at the start of this thread but clearly not.

 

 

[jabbr]

34 minutes ago, ecwl said:

All of you are right. I don't know how Pulse Array DAC works. And Rob Watts only "vaguely" describes it on the Head-Fi forum. But of course, why should he have to share it if most other DAC designers don't share every exquisite detail? Everybody has their secret sauce whether it's for real or not.

 

Anyway, this is my speculation of how it works. But I'm pretty sure I'm wrong in some ways. I thought most of you who have been commenting have read this already since it's at the start of this thread but clearly not.

 

 

 

 

So you are clearly missing the entire point of my last post... since you seem to be interested in speculating about how it might work you might want to take my comment at face value. ?

[mansr]

13 hours ago, The Computer Audiophile said:

Is there an app that does offline conversion using the same filters?

If you know the coefficients, there are many applications that will do the filtering.

[Whitigir]

2 hours ago, mansr said:

If you know the coefficients, there are many applications that will do the filtering.

Filtering is ok, but what about that specific WAT filtering ? That is my main question 

[mansr]

11 minutes ago, Whitigir said:

Filtering is ok, but what about that specific WAT filtering ? That is my main question 

WTA is just his made-up name for linear phase.

[Whitigir]

Now, only if HQplayer was able to do offline conversion too !

[rayl1234]

6 hours ago, mansr said:

WTA is just his made-up name for linear phase.

 

I read it differently.  I read it as the particular set of coefficients that achieve a certain error bound but are computable on the particular FPGA's capacity -- computable in the sense that unlike math, you don't have infinite precision...  The term "better than 16 bits" is thrown around, but unfortunately, that is one of the areas where I've not gotten a clear answer.  Is the implication that:

 

1. Each coefficient is w/in 2^-16 of the coefficient of the mathematical 1MM windowed sinc?  (This would say very little interesting as certain coefficients accumulate much greater errors than others, obviously)

2. Compared to using a mathematical 1MM windowed sinc, the results computed will match the mathematical computation when quantized to 16 bits?

3. For all bandwidth limited signals, the result of applying the filter will result in the equivalent digital values as a 705.6kHz sampling of the analog signal. (This would be the strongest defition, but this has not been confirmed much less had a proof presented).

 

However, I think my original thesis still stands -- although technically "feasible" in software, I haven't seen a truly packaged solution even for offline conversion, much less for streaming. (I will say, not to bring up a big debate on the topic, that I have my reasons to be a bit skeptical of "5%" CPU bec my admittedly unoptimized offline test code when I tried it last fall was closer to 60% of a 4 core CPU... Of course, on a GPU with 2000+ DSP cores, it would be easy, but would probably consume around 70+ watts of power.)  Thus, if you like what WTA does for you (and I do), all other chatter is just noise until this situation changes.