AudioQuest adds MQA Support to Dragonflies via firmware

[miguelito]

Interesting discussion on Audiostream.com today.

 

Apparently, there are 1) MQA "renderers", which require the player to do the first "unfolding", whereupon the renderer will do the final processing and 2) "decoders", which will play MQA files without any additional software required.

 

The DragonFly will apparently be a "renderer", not a "decoder" - which means the only software it will work properly with for MQA playback at this time is Tidal.

 

Not something I understood at all until today - maybe others were already aware.

Can someone with an Explorer2 do the following experiment please?

 

Play an MQA file with passthrough selected - the blue light should light up.

 

Does the blue light come on when you set Tidal to decode?

 

I am guessing the answer is no.

[miguelito]

Actually, according to the article at Audiostream, the answer s/b "Yes". i.e., for a full decoder, the first "unfold" can be in software without disrupting the ability of the decoder to perform the rest of the decoding. Or it can be direct to the DAC.

 

Will be interesting to see the answer to this one, though.

Precisely why I am asking.

[miguelito]

I guess it wasn't possible to squeeze the MQA decoder onto the puny microcontroller in those DACs.

Or... They are not giving up on some "mistique"...

[miguelito]

Yep, I got that strange clicking noise trying to use mine with the smaller Lightning to USB camera adapter and an iPhone 7. Works fine with my old iPhone 5s, though.

Same issue on iPhone 7+. The USB3 adapter fixed it. It's bigger, but at least it allows you to listen and charge at the same time.

[miguelito]

It's an 80 MHz CPU with 32 kB RAM and 512 kB flash, best case (I can't find the exact chip they use). I doubt that's enough to run the decoder.

Do you think it might be the very last bit, ie tailoring to the DAC, that's being done in the DragonFly?

[miguelito]

Maybe it just turns on the blue LED. Who knows?

Funny to be using bandwidth to do just that. Just as Bob Stuart intended it.

[miguelito]

When the update arrives and someone measures the sample rate at the DAC pins.

Or the analog output bandwidth...

[miguelito]

LAME!

[miguelito]

53 minutes ago, DarwinOSX said:

 

I wonder how this compares to the Meridian Explorer 2.  Will the Dragonfly do the full unfold?

As @mansr said, no it won't. This means it will only render MQA on desktops currently, either via playback of an MQA file via Audirvana or an MQA stream via either Audirvana or the native TIDAL app. Once the iOS/Android TIDAL apps allow unfold (which surely has to be coming soon) then you could use the DF on those mobile devices to render the MQA data.

 

It would be interesting to see if anyone who UNINTENDEDLY captured the unfolded MQA stream out of the TIDAL app, and INADVERTEDLY transferred the captured file to an iOS device only to FORTUITOUSLY use the Onkyo HF Player app to play back said file through an INTENTIONALLY updated DF would see it render the MQA file appropriately.

 

Which brings up the question... Is there an indicator, such as the colored sample rates on the DF (which are super useful), that would indicate MQA rendering is ON?

[miguelito]

32 minutes ago, jhwalker said:

 

As far as I can tell, no   It only uses the same lights already defined:  

 

Red: Standby Green: 44100.0 Hz Blue: 48000.0 Hz
Amber: 88200.0 Hz Magenta: 96000.0 Hz 

 

I'll be spending some time later today to see how it works with the various desktop playback sources (e.g., Tidal, Audirvana, Roon, etc.) on known MQA material and at various sample rates.

Rendering will have to be to a higher bitrate than 96 I would think. This is in principle possible as the DAC chip itself can go to 384 - AFAIK it is the controller that is the source of the limit. My understanding is that the rendering involves setting upsampling parameters (extracted from the unfolded MQA PCM stream) for the DAC to use.

 

The only apps that can produce an unfolded PCM stream are Audirvana and TIDAL at the moment, so Roon is of no use in decoding MQA at the moment. I suppose you could be a Roon beta tester in which case maybe you have a beta version of Roon with the unfold lib built in...

 

[miguelito]

1 hour ago, mansr said:

Judging by the specs of the microcontroller in the DF, I estimate that it is too slow to do even the rendering part. Therefore I suspect all it does is set the upsampling filter coefficients in the ESS DAC chip according to the parameters indicated by the input stream.

Yes, that was my expectation as well.

[miguelito]

15 minutes ago, jhwalker said:

Here's a screenshot of Audirvana playing with an upgraded DragonFly Red (firmware 1.06) - clearly shows input and output both at 352.8:

 

 

Fabulous!

Lossy music at 352KHz!!! (I could not help the bitchy comment... )...

 

BTW... Since Audirvana is a mac program, are people on Windoze not able to unfold MQA files?

[miguelito]

1 hour ago, Wavelength said:

DragonFly only supports sample rates over USB up to 96K, therefore DragonFly does not do the full unfold. Therefore applications such as Tidal and Audirvana are currently the only applications that support MQA and DragonFly. Expect Amara to introduce products next month and expanded Audirvana soon.

Can you explain a bit what happens in the DF once it receives the unfolded PCM stream? I see you replied to mansr so I am wondering what it actually is.

 

Quote

The passthrough in Tidal will just send the MQA file as is to the DragonFly. You must have Exclusive mode checked for MQA to work with Tidal and DragonFly.

If you choose pass-through, then TIDAL won't be doing the first unfold I don't think - unless it still does it because it detects a DAC that has MQA rendering capabilities?

 

Quote

All of you should give Audirvana 3.0 a try as well.

Yeap, A+ 3 is fabulous.

[miguelito]

1 minute ago, Wavelength said:

 

If you choose pass through then the MQA is basically not enabled.

 

Why do people think this is all about upsampling and decompression.

 

Guys if the files were compressed and require decompressing in a certain format then how do NON-MQA DACS playback MQA files?

 

Look it's much more than what everyone is speculating about. This is probably part of the problem with companies who think MQA is a bad thing. Maybe not, they might be a lot smarter than I am. But as a musician for some 50 or more years, I can tell you this is the real deal. Lowering the noise floor is a real undertaking.

 

Anyway, before you pass judgement you should listen.

 

Thanks,

Gordon

 

I've listened and I love what I heard.

[miguelito]

2 minutes ago, citsur86 said:

I see!  Do you happen to know of any of Tidal's songs offhand that are 24/192?  I'd like to try some out.  Also, how would I even know if I was listening to 192,000Hz if the Audio MIDI Setup only goes to 96,000Hz?

 

Audio midi will not know anything about MQA. It only displays the raw PCM rate to the device - it will never be higher than 96KHz for a Dragonfly (the current ones obviously).

 

Example: https://tidal.com/album/2400318

[miguelito]

11 hours ago, plissken said:

MQA doesn't solve a problem and it certainly isn't going to transcode and fix poor PCM or Tape Masters.

In my view MQA serves a sort of "quality remaster" tag... In the many examples I have heard, I have the impression that most of the improvement in quality comes from careful remastering rather than any technology improvements in MQA itself. I'm totally fine with that.

[miguelito]

23 hours ago, ShawnC said:

Unfortunately with the Dragonfly it's hard to figure out what exactly your getting.  It would help if Tidal listed what it was sending you if you could unfold it to it's maximum potential.  I think something like Audirvana shows this and the Mytech Brooklyn DAC does this too.

If Audio MIDI says that DF is operating at 88 or 96, then the MQA file has been unfolded and the DF should render the file properly. However, you don't know what the final resolution will be - it depends on the original MQA file.

 

Audirvana reads the MQA metadata properly and will spell it like it is, telling you exactly what is happening.

[miguelito]

1 hour ago, Chel2772 said:

I think it has to do with my internet speed.  If it's too slow it won't stream the MQA. but if I let it load, and then play the light goes purple.  Downside of intrenet in central Africa...

Nah. If you pick an MQA file, for example:

 

https://tidal.com/album/2400318

 

You should be able to play properly. There's no downsampling happening, and there's no "switching to non-MQA version when bandwidth is not enough" as far as I know.

[miguelito]

36 minutes ago, CNoblet said:

After you've gone to TIDAL Settings, have selected the Streaming header, and then selected the Dragonfly Red DAC, hover your cursor over the DAC's name and a gear wheel will appear.  Click on the gear wheel and another menu will appear.  Select "Exclusive" mode and you should be in business.  

Correct. DO NOT select "MQA pass-through" as in principle this should not unfold, which is required for the DF. I can't recall if TIDAL will actually ignore this setting with a DF, but just to be sure don't choose that.

 

36 minutes ago, CNoblet said:

If you have a mac, don't select "Force Full Volume" unless you want to damage your headphones. 

This is correct, but it needs a bit of explanation as in most cases people are instructed to use this setting.

 

For many (most?) DACs, the volume control on the mac/pc will change the PCM stream by applying the volume reduction in software before it reaches the DAC. This will destroy the MQA encoding. Enabling "Force Full Volume" guarantees the PCM stream will not be altered, but obviously will require that you have a volume control at the DAC or preamp.

 

For the DF, the computer's volume control actually controls the DAC itself, it does not change the PCM stream reaching the DAC, just sends instructions over USB on what volume to set the DF at (*). This is actually a great feature of the DF in my opinion.

 

(*) The DF Black actually has a USB-controlled analog volume control (applied on the final analog stage) whereas the Red does this digitally in the DAC chip.

[miguelito]

2 hours ago, mansr said:

Isn't that what I've been saying all along?

Yes it is. And as much as I respect Gordon Rankin, he needs to be a little less dogmatic here. For example, someone used the term decompress to refer to embedded information in the pcm stream to set such filters. And no, there is no real information other than vagaries and mumbling in MQA Co's pages.

 

I would like to understand how filters are tagged and selected. The MQA file must have a universal dictionary of these than the specific DAC implementations translate into the actual filters for each DAC API.

[miguelito]

1 hour ago, abrxx said:

Well Gordon did say that the microcontroller is doing DSP as well.

Is the microcontroller doing the DSP or setting DSP params in the ESS chip? Not the same. 

[miguelito]

22 minutes ago, abrxx said:

 

I am software developer not a hardware guy but it seems to me highly implausible that custom assembly code had to be written by the micro-controller's manufacturer purely to set ESS filters. Gordon has kindly divulged this info, so I can only assume that certain DSP is being done by the controller, as well as the setting of the ESS filters. As to what DSP that might be doing, I suggest we all go back and re-read the patent on how to implement the second and third folds. I posted the link to this is another thread.

 

One thing that is not clear to me is this issue of lowering the noise floor. Exactly what part of the MQA process is responsible for this? Does avoiding un-necessary sample rate conversions give us a better noise floor?

I am not suggesting ANY custom assembly code of any kind. What I am suggesting is that the MQA pcm data has information embedded that chooses a particular filter type. All the controller needs to do is have a table that maps that generic MQA specification to a setting in the DAC. 

 

Example: the MQA data might say "MIN PHASE ORDER 7". The controller translates that to ESS speak, which for this might be "O7 MP", as it hands over the pcm stream to the ESS DAC. 

[miguelito]

3 minutes ago, Wavelength said:

miguelito,

 

First decompressed was used out of context. Tidal downloads in FLAC you don't decompress FLAC, you uncompress it. This was the wrong term and confused several people.

 

I don't know exactly how MQA determines which filter to use or how to unfold the data. The USB stream has some identifier in it that tells the MQA library in the DragonFly that it is MQA and what filter to load into the ESS DAC chip and also how to unfold the data DSP and other things.

 

If any of you remember Pacific Microsystems HDCD format what they did was embed in the LSB a signature for their custom filter and output level. MQA maybe doing the same thing.

 

Really why do we need to know details like this? Why don't you just enjoy what we spent months creating for you.

 

Thanks,

Gordon

Thank you.

 

As Richard Feynman once said: "Understanding the science behind a flower makes me appreciate its beauty more, not less."

 

 

[miguelito]

14 hours ago, ShawnC said:

Doesn't MQA have the right to it's trade secrets or patents?  Wouldn't that be the competitors quest to figure out what's going on and try to make it better/cheaper.  Just asking.  

It's my prerogative to ask as much as I wish and conjecture as much as I feel like about anything. That's called freedom and the scientific enquiry. Yes, of course MQA has a right to their protected IP. I have every right to backward engineer that. I don't have the right to market a lossy compression mechanism that can be proven to be identical to MQA. 

 

I also have the right to raise my voice and complain when there is deceitful advertising, when I see that a technology could result in us being worse of on the whole at the expense of someone getting license money, etc.

 

So don't question the freedom we have to ask, conjecture, and voice concerns.

 

Thank you.

[miguelito]

Can someone with enough knowledge say whether this conjecture is at all correct?

 

My impression with unfolding is that it effectively accomplishes something similar to DSD. If I recall correctly, DSD dynamic range or effective resolution decreases at higher frequencies. I think this is easy to explain if we think of the high speed of 1's and 0's determine the waveform going up or down by a quanto: the higher the audio frequency, them coarser this process can work. So this means DSD has actually much higher resolution than PCM for lower frequencies, and lower for higher frequencies, which in fact is what is argued to be the case for human hearing. It seems to me that MQA's folding/unfolding is actually doing a very similar thing. Hence the lower noise floor in critical parts of the spectrum that Gordon is talking about.

 

Is there any merit to this interpretation?

 

Thx.