Connect no longer bit-perfect?

Do I think hi res is better?
Not to the point where I have bothered with it.

Now that I have read a little more about MQA, the quoted words are interesting.
Because on MQA, I find this essence:
"MQA attempts to approach the quality of the higher resolution file while using a compression technique that is closer to the CD end of things."
If so, it goes without saying that MQA cannot be better than the higher resolution file that it has compressed. At best it can be equal, and at anything less than best, not as good.
How does that then square with your finding MQA to be a great leap over Hi Res?
Could it then just be the old variable that we know so well - mastering? And MQA just a new bottle for the old red herring of Hi Res?
Any normal Hi res file can be converted to 16/44 to yield the same sound via 16/44 kit. If this can also be done for the MQA files that have moved you to "tears", see what happens when you play these via Sonos.

I have found that MQA exceeds HiRes on most files.
I couldn't work out why it often sounded so much better, and started to look into it more, it is not about resolution, it is about encoding the file digitally without loosing out on timing detail, almost looking back at why analogue can sound so good.

I think the soundbites that say "Hi Res quality from a small file" are more confusing that good.

This was the best explanation of why MQA works so well.......


MQA part 1; Why 24 bit 192 kHz audio?
https://www.youtube.com/watch?v=r_wxRGiBoJg

MQA part 2: how does MQA work
https://www.youtube.com/watch?v=T5o6XHVK2HA&t=56s


They are a bit long, but worth a watch.

Remember, Bob Stuart has always said anything above 24/96 is pointless, he changed his mind. Not saying higher res is needed for playback, but for encoding it adds a lot.


http://www.computeraudiophile.com/content/694-comprehensive-q-mqa-s-bob-stuart/

http://www.computeraudiophile.com/content/694-comprehensive-q-mqa-s-bob-stuart/
I am not an engineer, so my eyes glazed over very early with all the incomprehensible jargon. I persisted but I saw no point once I saw the graph that places CD at a lower sound quality capability than an LP. And you have to accept that Stuart has money to make in pushing his view.
If you or anyone is able to write about MQA being distinguished as different in a single variable double blind test of the kind that no one in the world of real science has a quibble with, it would be of interest. The next question then would be that which of the two different sounds are closer to the performance, which is more HiFi. Different does not automatically place it on the side of HigherFi.

That's the thing, MQA cannot be downsampled to standard resolution, Meridian won't allow it. They specifically forbid any of their licensed MQA products from outputting raw digital data. Therefore it is impossible to confirm how much is the higher resolution (science says the answer is none) and how much is the special mastering done by Meridian (science says 100%) via an ABX test. My, how convenient for Meridian is that? :8

What one could do is strip any hypersonics out of the analog MQA output and resample to 16/44.1, but I highly doubt any audiophile would call that a valid test. Though if you look at the MQA specs, the only thing it accomplishes is stripping extra bits from the bit depth and dedicating them to frequencies above 22 KHz.

Therefore it is impossible to confirm how much is the higher resolution (science says the answer is none) and how much is the special mastering done by Meridian (science says 100%) via an ABX test. My, how convenient for Meridian is that? :8

Yes, it is a clever copout. As cover for a little more sophisticated snake oil, it looks like.
To me, the controlled ABX listening test is Occam's razor to be used every time some one puts together graphs and jargon for the next best thing. Based on a very useful question attributed to the late Dudley Harwood, the founder of Harbeth speakers. "That's all very well laddie, but does it sound any different/better?" was his response to any suggested improvement in speaker design. Sound - pun intended - common sense.

At the same time, expect challenges if you make statements that Sonos isn't meant for high end audiophile listening, because that misleads existing and potential customers........

Of course, leading people to believe that it is capable of true high end performance, whilst failing to provide any ABX double blind testing to prove it, is equally misleading, surely... And insisting that people prove any claims (e.g. HiRes, MQA) with proper scientific testing, yet letting absurd claims like this go completely unchallenged, must surely be the ultimate in hypocrisy...

Sonos is very good at what it does, at it's price point.... Continually insisting that it's something it isn't can only lead to disappointment.

whilst failing to provide any ABX double blind testing to prove it

What a great line to make a last response to you, for anyone else that chances on this!

I believe that the onus for this lies on those that hear differences, not on those that do not, because ABX testing is used to establish that a difference exists. An ABX test, when the result is positive, is a statistical proof that a difference is audible. That is the main use of ABX testing.

When the result is negative, it is more complex to analyse, and a negative ABX doesn't prove anything. It doesn't prove that there is no difference.

Continually then asking for this to prove no difference between Sonos and High End - whatever that imprecise phrase means - isn't hypocrisy, it is just silly and childish and points to a lack of understanding of ABX testing. Particularly when I have also often said that speaker selection is a subjective matter of preferences between one good and another just as good, best done by listening to it in the kinds of environments that it is designed for, and when it is set up to deliver all of its designed performance. Someone that is disappointed after following that advice isn't anything I lose sleep over.

And I have more than once set out exactly how I did a blind level matched test - not double blind ABX - of two DACs of relevance to me, to see if I hear differences. Saying again what I did there - if I did not hear a difference in this test, it was very unlikely that I would hear it if the full ABX protocol was followed in this case. Did I prove that no difference exists? Of course not, and I did not say so then as I do not say so now. But I believe I have done more than most here that just spout claims to have heard differences that are on the superior side of sound quality.

Continually then asking for this to prove no difference between Sonos and High End....

I'm flexible - I'm happy to take advice on an agreed scientific method to evaluate the situation. You continue to provide no evidence whatsoever that Sonos kit can compete with high-end audio kit (let's cap it at $40k, shall we, for obvious reasons) yet continue to make these misleading statements. Whatever their CEO says in interviews, Sonos know better than to make such ludicrous claims in their advertising, as they would soon be caught out. Do you really think that, if it were even remotely true, that they would fail to capitalize on it?

And I have more than once set out exactly how I did a blind level matched test - not double blind ABX - of two DACs of relevance to me, to see if I hear differences.

Which has no relevance whatsoever when we're discussing whether Sonos can audibly compete with high end kit.

In ABX testing (and double blind testing inn general), the default position (the null hypothesis) is that A & B are indistinguishable from each other. That is the starting point and the assumption that should be made going into the test. Any test which approaches this from a different point of view (e.g. with the assumption that A & B are different) is inherently broken and bogus.

By definition the default position, the null hypothesis, the assumption that A&B are equal, cannot be proven. It can only be disproven. That is why it is chosen as the null hypothesis as we are looking to see if there are significant statistical deviations which show it isn't valid.

Note that ABX tests do not give a black and white, binary result. They give statistical output which can be used to judge how significant the deviation from the null hypothesis is; how likely it is that the null hypothesis has been invalidated.

Of course, in extreme examples (for instance, an ABX comparing an AM radio to a high-end hifi) will give clearly statistical evidence that they are not equal, that the null hypothesis is disproven.

However, that's not the sort of test we are talking about here. We are talking about differences that, if they exist, are extremely subtle and are likely to suffer from a lot of noise and error (statistically speaking).

So in summary:

The default position is that sources are the same. That position is impossible to prove.

Demanding so is unreasonable and unscientific. The onus is on people who claim otherwise to demonstrate a significant statistical deviation from that position.

Or, put another way, "extraordinary claims demand extraordinary proof".

The ball is in your court.

Cheers,

Keith

@Peter, if you are still here! - circling back to topic, this from what Sonos still claims for the Connect:
"CONNECT streams the music you love without loss of fidelity and at full CD-quality bandwidth".
Any comment? Ludicrous claim?:D
By the way, they also say this about the 5 unit:
"Attention obsessive audiophiles and hard-core music lovers. PLAY:5 is your speaker."
I can't speak to that bit because I am not an obsessive audiophile, I have outgrown that phase🆒