MQA is Vaporware

[Ishmael Slapowitz]

Jim Austin:

 

"When I played an MQA file through each of two grouped zones to both MQA-enabled DACs, both indicated that they were decoding MQA, which, according to both Roon and MQA experts I talked to, is a clear indication of bit-perfect playback. Indeed, dCS writes in the Bartók manual, "MQA decoding is not possible if the original MQA data has been changed." 

 

https://www.stereophile.com/content/dcs-bartok-da-processorheadphone-amplifier-page-2#QFfvFtGjlKGVc3zS.99

 

???? @mansr

[Samuel T Cogley]

12 minutes ago, Ishmael Slapowitz said:

Jim Austin:

 

"When I played an MQA file through each of two grouped zones to both MQA-enabled DACs, both indicated that they were decoding MQA, which, according to both Roon and MQA experts I talked to, is a clear indication of bit-perfect playback. Indeed, dCS writes in the Bartók manual, "MQA decoding is not possible if the original MQA data has been changed." 

 

https://www.stereophile.com/content/dcs-bartok-da-processorheadphone-amplifier-page-2#QFfvFtGjlKGVc3zS.99

 

???? @mansr

 

 

I thought it was @FredericV who showed that changing certain bits of the payload would alter the sound, but the blue light would remain on (MQA's "authentication" effectively circumvented).  Apologies if I'm mistaken.

[The Computer Audiophile]

3 minutes ago, Samuel T Cogley said:

 

 

I thought it was @FredericV who showed that changing certain bits of the payload would alter the sound, but the blue light would remain on (MQA's "authentication" effectively circumvented).  Apologies if I'm mistaken.

Sure, but that's a science project that no end user will undertake just to see if the authentication light will still illuminate.

[Ran]

14 minutes ago, The Computer Audiophile said:

Sure, but that's a science project that no end user will undertake just to see if the authentication light will still illuminate.

 

Still does not justify providing erroneous information to the end user.

[mansr]

25 minutes ago, Ishmael Slapowitz said:

Jim Austin:

 

"When I played an MQA file through each of two grouped zones to both MQA-enabled DACs, both indicated that they were decoding MQA, which, according to both Roon and MQA experts I talked to, is a clear indication of bit-perfect playback. Indeed, dCS writes in the Bartók manual, "MQA decoding is not possible if the original MQA data has been changed." 

 

https://www.stereophile.com/content/dcs-bartok-da-processorheadphone-amplifier-page-2#QFfvFtGjlKGVc3zS.99

 

???? @mansr

The low 8 bits can be altered without the blue light going out. It is unlikely that something would mess with only these bits while leaving the higher ones unchanged. This makes it reasonable as a quick check for bit-perfect transport.

[The Computer Audiophile]

Just now, Ran said:

 

Still does not justify providing erroneous information to the end user.

I agree for the most part, but almost any technology can be altered by someone in a lab, thus making the advertised "benefit" untrue. 

[Ishmael Slapowitz]

19 minutes ago, Samuel T Cogley said:

 

 

I thought it was @FredericV who showed that changing certain bits of the payload would alter the sound, but the blue light would remain on (MQA's "authentication" effectively circumvented).  Apologies if I'm mistaken.

You might be correct about this...it may well have been @FredericV

[Ishmael Slapowitz]

1 minute ago, The Computer Audiophile said:

I agree for the most part, but almost any technology can be altered by someone in a lab, thus making the advertised "benefit" untrue. 

that is the the point...Jim Austin, the editor of Stereophile, put something in print that is simply untrue. And how does that serve the consumer?

[The Computer Audiophile]

2 minutes ago, Ishmael Slapowitz said:

that is the the point...Jim Austin, the editor of Stereophile, put something in print that is simply untrue. And how does that serve the consumer?

If you take your statement to the extreme, almost any technology can be deemed a hoax.

 

If I setup a computer to capture packets outside your house and make your 4K TV streams 3.999K streams, you can tell Netflix they are lying because the streams aren't always 4K. Since technically this is possible, you better start telling Netflix to stop advertising 4K content because someone could alter the stream. 

[Samuel T Cogley]

6 minutes ago, The Computer Audiophile said:

I agree for the most part, but almost any technology can be altered by someone in a lab, thus making the advertised "benefit" untrue. 

 

It can be objectively stated that the "authentication" can be easily circumvented.  The practicality of this is not relevant to the fact that the "authentication" is weak.

[KeenObserver]

44 minutes ago, Ishmael Slapowitz said:

Jim Austin:

 

"When I played an MQA file through each of two grouped zones to both MQA-enabled DACs, both indicated that they were decoding MQA, which, according to both Roon and MQA experts I talked to, is a clear indication of bit-perfect playback. Indeed, dCS writes in the Bartók manual, "MQA decoding is not possible if the original MQA data has been changed." 

 

https://www.stereophile.com/content/dcs-bartok-da-processorheadphone-amplifier-page-2#QFfvFtGjlKGVc3zS.99

 

???? @mansr

 

Just can't stop shilling for MQA.

[Ishmael Slapowitz]

2 minutes ago, The Computer Audiophile said:

If you take your statement to the extreme, almost any technology can be deemed a hoax.

 

If I setup a computer to capture packets outside your house and make your 4K TV streams 3.999K streams, you can tell Netflix they are lying because the streams aren't always 4K. Since technically this is possible, you better start telling Netflix to stop advertising 4K content because someone could alter the stream. 

that is picking nits...Austin quotes dCS saying that if the "data has been changed" no blue light. 

[The Computer Audiophile]

2 minutes ago, Samuel T Cogley said:

 

It can be objectively stated that the "authentication" can be easily circumvented.  The practicality of this is not relevant to the fact that the "authentication" is weak.

I can also easily change your 4K tv stream to 3.999K, so Netflix should stop advertising 4K?

[The Computer Audiophile]

Just now, Ishmael Slapowitz said:

that is picking nits...Austin quotes dCS saying that if the "data has been changed" no blue light. 

Tell me a practical, non-lab, scenario where this could occur. 

[Ishmael Slapowitz]

Just now, The Computer Audiophile said:

I can also easily change your 4K tv stream to 3.999K, so Netflix should stop advertising 4K?

that is plain silly, because no human on earth will know the difference. 

[Samuel T Cogley]

Just now, The Computer Audiophile said:

I can also easily change your 4K tv stream to 3.999K, so Netflix should stop advertising 4K?

 

Non sequitur.  Netflix isn't advertising "authentication" of the stream.

[Ishmael Slapowitz]

Just now, The Computer Audiophile said:

Tell me a practical, non-lab, scenario where this could occur. 

how about the fact that dcS and did not even know how Roon works with a multi zone set up? 

[Samuel T Cogley]

4 minutes ago, The Computer Audiophile said:

Tell me a practical, non-lab, scenario where this could occur. 

 

Doesn't matter.  The authentication has been broken.

 

EDIT: If you alter the file, and the blue light still comes on, the "authentication" is bogus.

[The Computer Audiophile]

13 minutes ago, Samuel T Cogley said:

 

Doesn't matter.  The authentication has been broken.

 

EDIT: If you alter the file, and the blue light still comes on, the "authentication" is bogus.

You better call Dolby as well because the same thing can be done for DTS etc...

[Samuel T Cogley]

6 minutes ago, The Computer Audiophile said:

You better call Dolby as well because the same thing can be done for DTS etc...

 

Don't recall Dolby heralding their "authentication" scheme.  Detecting encoded material (even if it's intentionally adulterated) is a far different thing than advertising your "secure authentication" as a guarantee of "what the artist intended", only to demonstrate that the "authentication" light will still illuminate when the file has been altered.  Apples and oranges.

[The Computer Audiophile]

1 minute ago, Samuel T Cogley said:

 

Don't recall Dolby heralding their "authentication" scheme.  Detecting encoded material (even if it's intentionally adulterated) is a far different thing than advertising your "secure authentication" as a guarantee of "what the artist intended", only to demonstrate that the "authentication" light will still illuminate when the file has been altered.  Apples and oranges.

I think attacking MQA based on the fact that someone in a lab can change the file and make the blue light shine even though the file has been altered is rather dumb. 

 

I guarantee that you're browsing this site via SSL and your communication is encrypted if you see the pad lock in your browser. All people saying SSL is secure and people who show the padlock should immediately cease saying this is secure because this can be defeated in a laboratory. 

 

Makes no sense to me. 

[Samuel T Cogley]

Just now, The Computer Audiophile said:

I think attacking MQA based on the fact that someone in a lab can change the file and make the blue light shine even though the file has been altered is rather dumb. 

 

I guarantee that you're browsing this site via SSL and your communication is encrypted if you see the pad lock in your browser. All people saying SSL is secure and people who show the padlock should immediately cease saying this is secure because this can be defeated in a laboratory. 

 

Makes no sense to me. 

 

Assuming you're not revealing an SSL vulnerability with your site(s), change just one bit of the data in the encrypted stream, and the alteration detection mechanisms will do their job and you'll know something's amiss.  And by the way, editing binary data in a file does not require a "laboratory" or even a white coat.  🙂

 

And I'm not "attacking" MQA.  I'm simply stating that the "authentication" has been effectively circumvented.  This is a fact.

[The Computer Audiophile]

1 minute ago, Samuel T Cogley said:

Assuming you're not revealing an SSL vulnerability with your site(s), change just one bit of the data in the encrypted stream, and the alteration detection mechanisms will do their job and you'll know something's amiss.

You must have missed Edward Snowden's data dump. The capability is there. 

[Samuel T Cogley]

5 minutes ago, The Computer Audiophile said:

You must have missed Edward Snowden's data dump. The capability is there. 

 

Those vulnerabilities are a few years old and have been addressed.  There's no doubt that SSL has its problems (not the least of which is the unknowns around possible backdoors in AES), but you can't compare it with the utterly broken "authentication" of MQA.  They're completely unrelated.

[The Computer Audiophile]

Just now, Samuel T Cogley said:

 

Those vulnerabilities are a few years old and have been addressed.  There's no doubt that SSL has its problems (not the least of which is the unknowns around possible backdoors in AES), but you can't compare it with the utterly broken "authentication" of MQA.  They're completely unrelated.

Um, OK.