MQA streaming detailed analysis --- and it's not good

[Eppie]

GoldenSound on YouTube has performed a detailed analysis of MQA on Tidal. It is a technical analysis that may be beyond the level of the average audiophile, but for those that understand encoding methods and in particular those with recording engineering experience, it's a very interesting video. The first half of the video contains the technical details and the second half some personal opinion and comments.

Short answer for audiophiles is stay with FLAC. What MQA does to audio streams is not lossless and in some cases downright detrimental. How they market their services and make it extremely difficult for independent analysis is much too similar to what some dubious peripheral manufacturers are doing and is great cause for concern. With Tidal replacing FLAC streams with MQA streams and defaulting to the MQA version, they may be doing themselves harm in the long run and their subscribers a disservice. The MQA files can even be larger than the FLAC files so it's not even about bandwidth and appears to be all about royalties. Fortunately there are good people out there willing to take the time and effort to dig deeper and reveal the facts about some of these technologies. There is also a lot of other material referenced to in the description if you want to dig deeper, and he linked to every section of the video so you can jump to the conclusion and comments if you want to skip the details.

(Edit)
@Irvrobinson linked to the AudioScienceReview discussion in another thread and I added that link for more info:
https://www.audiosciencereview.com/forum/index.php?threads/mqa-deep-dive-i-published-music-on-tidal-to-test-mqa.22549/

 

[Trebdp83]

Unfortunately, just another attempt at getting a few bucks out of participants for use of a process. And who can afford it anyway after dolby has picked everyone’s pockets? I do enjoy sampling atmos albums via Tidal, but will probably not renew my subscription after the three months for three bucks offer comes to an end.

 

[Pogre]

I've yet to find a streaming service that equals my cd rips, even Amazon HD falls a little short, imo. I'm just gonna stick with my cd rips.

 

[John Parks]

Response from Bob Stuart:

All that glitters is not gold(en) | Bob Talks

A Deeper Look into Music and Test signals

bobtalks.co.uk

 

[lovinthehd]

Bob Stuart has appropriate initials.

 

[Eppie]

Response from Bob Stuart:

All that glitters is not gold(en) | Bob Talks

A Deeper Look into Music and Test signals

bobtalks.co.uk

I have hard time believing someone that claims their technology provides "higher temporal resolution" and "lower blur" without any measurements to back up that claim or even a proper explanation of what that means. Temporal resolution is a listener's ability to follow rapid changes in sound. (There are studies for example that show that vocalists have higher temporal resolution than non-musicians and slightly higher than even other musicians.) How would that apply to a compression technology? Is it higher than with other technologies and if so how? How do you measure temporal resolution in compression codecs and above what resolution does it no longer matter? How can you have lower blur (horizontal smearing of decompressed wave forms) when flac is lossless with no blur? It's a lot of fancy language with no evidence to back it up.

 

[John Parks]

I have hard time believing someone that claims their technology provides "higher temporal resolution" and "lower blur" without any measurements to back up that claim or even a proper explanation of what that means. Temporal resolution is a listener's ability to follow rapid changes in sound. (There are studies for example that show that vocalists have higher temporal resolution than non-musicians and slightly higher than even other musicians.) How would that apply to a compression technology? Is it higher than with other technologies and if so how? How do you measure temporal resolution in compression codecs and above what resolution does it no longer matter? How can you have lower blur (horizontal smearing of decompressed wave forms) when flac is lossless with no blur? It's a lot of fancy language with no evidence to back it up.

I really do not have an opinion either way. What is obvious is that both sides have their own agenda and that's fine. I first became aware of the GoldenSound video on another site and what really caught that reviewers attention was where the video shows measurable differences between digital cables...

 

[mtrycrafts]

I really do not have an opinion either way. What is obvious is that both sides have their own agenda and that's fine. I first became aware of the GoldenSound video on another site and what really caught that reviewers attention was where the video shows measurable differences between digital cables...

What kind of differences? There is a measurable difference between 5ft cable and 6 ft cable. or a 6'-1" cable if test gear is good enough.
What parameters are different and how much? Does it exceed perceptible differences?

 

[John Parks]

What kind of differences? There is a measurable difference between 5ft cable and 6 ft cable. or a 6'-1" cable if test gear is good enough.
What parameters are different and how much? Does it exceed perceptible differences?

In the video, jitter was specifically mentioned.

 

[lovinthehd]

In the video, jitter was specifically mentioned.

In real life jitter is what sort of audible issue?

 

[Jerkface]

In real life jitter is what sort of audible issue?

Rather than try to explain, you can listen yourself. Someone kindly posted audio samples of them basically upsampling then downsampling various wave files while subjecting them to various kinds of sampling errors that would be present in jitter caused by a variety of factors.

Pro tip: The 1Khz test tones are probably most revelatory in terms of "what" it does, but the music files are probably more revelatory in terms of "how much it matters".

what does jitter ACTUALLY sound like?

In a previous post we had an encounter with jitter in audio. The talk about jitter is hard to miss these days. There are plenty of products on the mar...

www.sereneaudio.com

 

[lovinthehd]

Rather than try to explain, you can listen yourself. Someone kindly posted audio samples of them basically upsampling then downsampling various wave files while subjecting them to various kinds of sampling errors that would be present in jitter caused by a variety of factors.

Pro tip: The 1Khz test tones are probably most revelatory in terms of "what" it does, but the music files are probably more revelatory in terms of "how much it matters".

what does jitter ACTUALLY sound like?

In a previous post we had an encounter with jitter in audio. The talk about jitter is hard to miss these days. There are plenty of products on the mar...

www.sereneaudio.com

Oh please explain....

 

[Jerkface]

Oh please explain....

I listened to it on my laptop speakers, not my system. The test tones, you can hear what should be a continuous sound have a waver to it, like a singer's vibrato. The music, again, I listened to it on my laptop, not my big system, so I'm not going to make a prejudgment regarding what it does or doesn't do.

 

[lovinthehd]

I listened to it on my laptop speakers, not my system. The test tones, you can hear what should be a continuous sound have a waver to it, like a singer's vibrato. The music, again, I listened to it on my laptop, not my big system, so I'm not going to make a prejudgment regarding what it does or doesn't do.

That was at the 100x or?

 

[Jerkface]

That was at the 100x or?

Actually, I just went back and listened again - apparently his test tone itself had the warble in it. The 2 microsecond (about the most jitter you'll ever get in real-world applications) created a significant amount of white noise similar to tape hiss. But it definitely wasn't present in with the same amount of jitter in an actual musical source. Once again, this is on craptop built-in speakers, though.

 

[lovinthehd]

Actually, I just went back and listened again - apparently his test tone itself had the warble in it. The 2 microsecond (about the most jitter you'll ever get in real-world applications) created a significant amount of white noise similar to tape hiss.

Never been an issue IME

 

[Jerkface]

Never been an issue IME

I mean, anything that raises the noise floor is a potential issue. But it's hard for a cable to add enough jitter to ruin the rest of your signal chain.

 

[lovinthehd]

I mean, anything that raises the noise floor is a potential issue. But it's hard for a cable to add enough jitter to ruin the rest of your signal chain.

Explain how a cable would add jitter please.

 

[Jerkface]

Explain how a cable would add jitter please.

Jitter is errors in the bitstream that are compensated on the receiving end through quantization, which adds noise. It's not hard to imagine (or measure) how a poorly constructed cable could add errors through poor/non-existent shielding, ill-fitting terminals, etc.

We're on the same side here regarding $10K cables, please don't get me wrong here.

 

[lovinthehd]

Jitter is errors in the bitstream that are compensated on the receiving end through quantization, which adds noise. It's not hard to imagine (or measure) how a poorly constructed cable could add errors through poor/non-existent shielding, ill-fitting terminals, etc.

We're on the same side here regarding $10K cables, please don't get me wrong here.

Sorry, hard to determine just what you're on about. Please refer to audibility of "jitter'