Bob Stuart launches blog to explain MQA

[mansr]

45 minutes ago, Miska said:

Where's the AI? It's not a cool thing if it doesn't include some form of Artificial Intelligence. (what in the past was boringly called statistics and correlation)

Haven't they claimed to use AI to determine what ADC "correction" to apply?

[mansr]

52 minutes ago, kumakuma said:

You have fallen into the trap of believing that "any publicity is good publicity."

 

Ask Boeing if that's true.

Master Quality Augmentation System.

[mansr]

2 hours ago, ARQuint said:

Thanks. I'd never call myself a journalist. The term should be reserved for the likes of Edmund R. Murrow, Seymour Hersh, and Jimmy Olson. I'm just not in that league.

You forgot Lee Scoggins.

[mansr]

4 hours ago, Em2016 said:

I know Meridian were right there very early on in these 2, but were they the absolute first to release these 2, for sure?

Does it even matter? The pieces were all in place, making inevitable the emergence of such products from multiple manufacturers. Being first to market may give a short-term financial boost (not that it helped Meridian, if they were indeed first), nothing more.

[mansr]

3 hours ago, firedog said:

Andrew -

That's really a useless argument. Qualifications don't prove anything, on their own, about the validity of one's pronouncements. Especially when commercial interests are involved. The statements need to be evaluated, not just the source. 

Linus Pauling - a genius and Nobel Prize winner in chemistry - made lots of idiotic and wrong pronouncements about medicine and other fields after winning his prize. 
Lots of medical doctors and academics with PhD's get involved with money making schemes and back what are at most "doubtful" projects with little or no actual utility. And in some cases projects that are harmful to consumers. Their degrees don't make their pronouncements correct or worthy of a  faith based type belief in them. 

 

BS's past doesn't mean anything he says about his MQA money making project is true or should be accepted b/c he says it. The fact is, that neither he nor any other MQA supporter has managed to refute the technical critiques made of MQA here. Instead, they, like you, attack the source of the criticism and not the substance. 

It should be obvious also to you that he does spout a lot of doubtful or even deceptive marketing junk when talking about MQA. One example: Still referring to it as "lossless", which is ,simply put, a lie - unless you qualify it with a word like "perceptually". There are more similar examples that I won't go into here. 

https://rationalwiki.org/wiki/Nobel_disease

[mansr]

58 minutes ago, crenca said:

Can someone speak to the AES however.  Why is Bob S a member of the current working group around "Hi Res" for example?  Do they not police their own, or is AES simply a vehicle for yet-more industry promotion and back slapping?

 

edit:  In other words, does the AES have integrity, or is its purpose to serve its members and give them "awards" and "accomplishments" which help them $sell$ stuff?  Is Bob S an aberration, or is it full of such charletons?

I think you've figured it out.

[mansr]

19 minutes ago, Kal Rubinson said:

When I did my first trials with multichannel MQA,

There's multi-channel MQA?

[mansr]

10 hours ago, Kal Rubinson said:

Not loose in the wild.  Because of my interest in multichannel, I asked Bob Stuart whether MQA would support it (and also I asked about EQ).  He said that both would be coming in the future.  He also sent me a selection of multichannel MQA files, many of which were familiar to me.  These were the subject of my listening tests.

What DAC supports such files?

[mansr]

9 hours ago, esldude said:

Its not 100% clear, but the Stereophile testing would make you think you do have control of the filtering.  They claim hardware implementation of MQA so I'm not sure that means it is buried in the USB controller, though MQA is only available via USB.

If MQA is only supported over USB, it usually means the decoder is done on an XMOS USB chip.

[mansr]

4 minutes ago, Kal Rubinson said:

A stack of three Mytek Brooklyn DACs.

So the MQA is simply three separately coded stereo streams?

[mansr]

3 minutes ago, STC said:

I am guessing they are just like any other 5.1 multi channel tracks. 

"Any other" 5.1 audio isn't MQA.

[mansr]

2 minutes ago, STC said:

I should have kept my mouth shut since I don’t and never will use MQA but as a Mytek user that’s how you play multichannel tracks with Mytek DAC which is MQA enabled or whatever fancy name they are using. 

The question was specifically how multi-channel MQA is coded. If the file is split into 2-channel pairs, do you get three valid MQA files?

[mansr]

1 hour ago, Kal Rubinson said:

The original file is a single PCM file just like any other multichannel file.

The file, of course, must be separated into channels for playback, just like any other multichannel file.  Typically, that is done in a multichannel DAC but it can be done separately as, for example, with a miniDSP U-DIO8 which splits it into 3-4 S/PDIF or AES3 outputs.

It can be done in a processor such as with JRiver running on a Mac which recognizes the 3-4 Myteks as individual stereo devices but can combine them into a "virtual" multichannel device.

In other words, MQA is handled just like any PCM file because MQA is embedded into the PCM data and none of these processes see it (until it gets to a suitable DAC).

How to route multi-channel audio to several stereo DACs wasn't the question. I was asking only about how the MQA data is encoded. I'm guessing it's simply done in channel pairs using the same format as stereo MQA, given that it can be decoded on separate DACs. That does mean you need to pair the channels correctly, of course.

[mansr]

1 minute ago, Kal Rubinson said:

MQA data is always buried in each channel of the PCM and, regardless of whether it is stereo or multichannel. 

No, that's not how it works. In stereo files, the MQA data is split between the channels. One channel on its own is useless; it can't even be recognised as MQA.

[mansr]

9 hours ago, STC said:

 But that doesn’t explain how center and bass can be processed as a pair?

 

Interesting to know that MQA requires stereo or a pair. Isn’t it the same as what MP3 does by converting  to joint stereo to reduce the file size further?  That alone would sound slightly different from the original. 

Almost all codecs, lossless ones included, use channel decorrelation one way or another to improve the compression ratio.