MQA Deep Dive - I published music on tidal to test MQA

[tmtomh]

I have not done controlled testing of undecoded MQA versus original to know the sonic effect. Objectively it is a degradation.

Decoding MQA though is free both in Tidal app and in Roon. So the question is, does it matter? MQA says the process of adding MQA bits to baseline layer is reversible and lossless. If so, then we are getting more, not less with that addition if you decode the MQA. Which again, those of us who consume MQA do.

Has anyone who complained about the impact on undecoded streams done listening tests? If not, why not? You are making that a huge part of your argument. Why don't we have listening test data to back your complaints as far as impairment?

Note that I have done my complaining on the objective side in my youtube video.

Let's see a few ABX test outputs on who can pass it and what they are hearing different. Don't wait on me on that. You don't need me for any of this type of testing.

Decoding MQA is not free in Tidal. It requires subscription to their highest-priced tier.

 

[mansr]

Also, when we compare the file size of MQA and 24/48 PCM FLAC - a format with sufficient bit depth/noise floor and plenty of ultrasonic room for any decent filter - what is the file size savings of MQA compared with that?

Negative.

 

[voodooless]

Let’s compare apples to apples then? How many bits do we really need? MQA also does not use the full 24 bit for the audible part, so take your 24 bit source FLAC, chop off the noise, and then compress what is left. That will be between 17 an 19 bits roughly in most cases? Now compare the size to the MQA version with half the sampling rate? I’m pretty sure it will be much closer. I think someone already did this exercise a while ago?

 

[mansr]

Decoding MQA is not free in Tidal. It requires subscription to their highest-priced tier.

Does the new middle tier deliver MQA but block decoding? That's evil.

 

[ebslo]

and when MQA says they're better than lossless they are indeed better (not necessarily to the consumers) because they're smaller and easier to maintain on servers and give perceptually the same thing anyway.

I'm still unclear on this point. Could you please clarify the following?
1. Is MQA encoded to 24-bit/48kHz smaller than FLAC-encoded 24-bit/48kHz PCM?
2. Does MQA encoded to 24-bit/48kHz deliver perceptually the same, more, or less than FLAC-encoded 24-bit/48kHz PCM?

 

[amirm]

For 99% of its existence, MQA has said - falsely - that it is lossless. They only changed to "better than lossless" in the last week or, obviously in response to the discussion prompted by @GoldenOne 's video and posts, which both Stuart and Amir have dismissed largely with ad hominem attacks on GO as a "blogger."

He is a "blogger." MQA provided the clear explanation of what was wrong with his test before he went public with his test. He ignored that and ran off with his testing and anti-MQA arguments he had read online, personalized the whole thing with MQA being mafia ready to go after him, etc. just like a blogger would do to sensationalize the topic.

Even after Bob Stuart explained, in more detail, why he was wrong, he doubled down and did the same "bogger" thing of they are wrong with another video. He doesn't have the technical background to understand the answer Bob gave. He should have come here, post a link to Bob's answer and ask whether what he was saying was correct or not. Instead, he produced another video.

OP has superb skills when it comes to writing and voice over with that "BBC News" accent so folks believe what he says especially since what he says is what they want to hear. But this is a technical topic and not about video and audio presentation skills on youtube. You all decided to run with an expert witness that has no expertise in this area and made mistakes like not knowing a journal paper is peer reviewed.

Bob Stuart on the other hand, has eared his status in research community through his many publications that are highly referenced and important to our understanding of audio. Yet you dismiss him as snake oil salesman and such, and complain when someone puts OP in the bucket he is. If you care about proper respect, then you better start with providing that for Bob Stuart and Peter Craven way before you do for OP. That you do the reverse, takes out any credibility in the arguments you are making.

As to "ad hominem attacks," I guess you forgot these he post about me:

This looks like the upstanding citizen you are asking respect for? He goes on to imply I have a shell company and stealing money from the government. And how "I" designed Microsoft DRM.

I suggest backing off from this lest you want to be dragged into the same mud with him.

 

[mansr]

1. Is MQA encoded to 24-bit/48kHz smaller than FLAC-encoded 24-bit/48kHz PCM?

It can't be. It's mathematically impossible.

2. Does MQA encoded to 24-bit/48kHz deliver perceptually the same, more, or less than FLAC-encoded 24-bit/48kHz PCM?

Less. Again, by mathematical necessity.

 

[mansr]

He is a "blogger."

What does that make you?

 

[amirm]

Decoding MQA is not free in Tidal. It requires subscription to their highest-priced tier.

And what content do you have that is in MQA without that tier?

 

[amirm]

What does that make you?

A person with Reply ban power over you if you keep posting about your emotions than anything related to the topic.

 

[amirm]

I'm still unclear on this point. Could you please clarify the following?
1. Is MQA encoded to 24-bit/48kHz smaller than FLAC-encoded 24-bit/48kHz PCM?
2. Does MQA encoded to 24-bit/48kHz deliver perceptually the same, more, or less than FLAC-encoded 24-bit/48kHz PCM?

What was the source file in #1?

 

[mansr]

A person with Reply ban power over you if you keep posting about your emotions than anything related to the topic.

Bob's attack dog, then. Thanks for the confirmation.

 

[tmtomh]

Does the new middle tier deliver MQA but block decoding? That's evil.

Others in the thread have provided screen shots documenting that the new lossless non-high-res tier includes "M"-labeled aka MQA files. MQA decoding is not available in that tier.

To be clear, no one has claimed that all of Tidal's content in that tier is undecoded MQA. But there is documented evidence of many MQA tracks showing up in that tier.

 

[ebslo]

(emphasis mine)

The MQA version maintains the music content instead of all the ultrasonic noise and is able to achieve much better compression ratio. And contrary to what someone says, it is absolutely comparable to 176 kHz because there is nothing to encode above 88 kHz so it just does a resample to get the DAC to play at that sample rate.

Yes, but is it not equally comparable to the FLAC encoded 88kHz PCM for that same reason? I acknowledge you did not say otherwise, I'm just asking directly as a point of clarification.

 

[q3cpma]

If it's aiming to be "perceptually lossless", why not compare it with lossy codecs with the same goal, instead? Because it's trying to be inbetween lossless and lossy? Maybe lossywav or wavpack's lossy mode, then. Another problem that is solved by real lossless is that the "made for real music and not test tones" bullshit doesn't include electronic music that could have both an inexistent noise floor (digital silence + possible dither) and some "test tone" like components.

He is a "blogger." MQA provided the clear explanation of what was wrong with his test before he went public with his test. He ignored that and ran off with his testing and anti-MQA arguments he had read online, personalized the whole thing with MQA being mafia ready to go after him, etc. just like a blogger would do to sensationalize the topic.

Even after Bob Stuart explained, in more detail, why he was wrong, he doubled down and did the same "bogger" thing of they are wrong with another video. He doesn't have the technical background to understand the answer Bob gave. He should have come here, post a link to Bob's answer and ask whether what he was saying was correct or not. Instead, he produced another video.

OP has superb skills when it comes to writing and voice over with that "BBC News" accent so folks believe what he says especially since what he says is what they want to hear. But this is a technical topic and not about video and audio presentation skills on youtube. You all decided to run with an expert witness that has no expertise in this area and made mistakes like not knowing a journal paper is peer reviewed.

Bob Stuart on the other hand, has eared his status in research community through his many publications that are highly referenced and important to our understanding of audio. Yet you dismiss him as snake oil salesman and such, and complain when someone puts OP in the bucket he is. If you care about proper respect, then you better start with providing that for Bob Stuart and Peter Craven way before you do for OP. That you do the reverse, takes out any credibility in the arguments you are making.

As to "ad hominem attacks," I guess you forgot these he post about me:

View attachment 132757

This looks like the upstanding citizen you are asking respect for? He goes on to imply I have a shell company and stealing money from the government. And how "I" designed Microsoft DRM.

I suggest backing off from this lest you want to be dragged into the same mud with him.

I kinda agree that the OP lacked some knowledge, and he did ran way too far with his suspicions (even if they may be partly shared by others, at least as concern) but he nonetheless did something useful. But I can't fathom how you could call the scandalous marketing bullshit full of rhetoric tricks used in the blog post response as just an "explanation".

 

[KSTR]

Let’s compare apples to apples then? How many bits do we really need? MQA also does not use the full 24 bit for the audible part, so take your 24 bit source FLAC, chop off the noise, and then compress what is left. That will be between 17 an 19 bits roughly in most cases? Now compare the size to the MQA version with half the sampling rate? I’m pretty sure it will be much closer. I think someone already did this exercise a while ago?

The "how many bits at which sample rate" questions etc have been competely addressed by Bob Stuart in this paper. It does not deal with MQA directly but with the foundations for it -- the room available in a PCM coding channel for additional information, plus a wealth of other interesting things.

 

[amirm]

(emphasis mine)

Yes, but is it not equally comparable to the FLAC encoded 88kHz PCM for that same reason? I acknowledge you did not say otherwise, I'm just asking directly as a point of clarification.

I don't understand your question. MQA encodes all the music information it can identify above 22.1 kHz. So in 88 kHz sampling, it would find and encode music content to 44.1 kHz bandwidth. In that regard from musical information point of view, it is the same as FLAC at 88 kHz. It is also the same as 176 kHz because there is nothing else left to encode from 44 to 88 kHz.

 

[DimitryZ]

First off, welcome to the forum!

As to your analysis and argument here, I personally lack the technical expertise to evaluate it definitively. I will say that it certainly seems reasonable, and the idea that we must evaluate measurable sonic differences in the context of audibility is one that is routinely used and agreed to here - in other words a measurable difference between two things (pieces of equipment, codecs, whatever) is not significant or important if we can be confident that the measured difference is below the level of audibility.

Where your argument runs into a problem, though, is when we come to the next question: Let's say MQA is audibly indistinguishable from conventional PCM. I am not ready to concede that definitively, but let's concede it temporarily for the sake of discussion.

If MQA is sonically identical for all intents and purposes, then two questions naturally present themselves:

1. Is there any non-sonic reason to prefer MQA over other formats? In other words, do we need MQA?
2. Is there any non-sonic reason that MQA might actually be actively detrimental? In other words, does MQA create problems for music consumers that could be effectively addressed if MQA disappeared?

To the first question, your analysis clearly shows that the answer is No - we don't need MQA. It solves a problem that doesn't exist, it produces none of the audible benefits MQA (the company) claims for it, and so on.

After we have answered question #1, we are essentially at a point where we share @amirm 's position: MQA is of no great consequence sonically and of no great consequence non-sonically - it's just another codec/format, so why bother worrying about it or trying to get people to come down strongly for or against it?

The problem is question #2, to which the answer is clearly Yes. There are several reasons that MQA's existence and presence in the marketplace can be seen as not simply neutral or unnecessary, but actively negative and problematic:

• It pollutes the audiophile streaming pool. It has been shown that Tidal's lossless, non-high-res tier includes MQA files, which in that subscription tier cannot be decoded. And whatever one thinks about the audible performance of unfolded or fully rendered MQA content, undecoded MQA content plays back at a bit depth that is inferior not only to 16-bit redbook, but based on the analysis of @mansr, @Archimago and others, apparently even worse than Philips' original 14-bit proposal for the format. With 13 bits of effective depth, undecoded MQA is basically taking things down to the level of high quality analogue reel to reel tape (albeit with better speed accuracy of course). This makes a complete mockery of the entire purpose of this site - if you are regularly playing 13-bit content, then who cares if a DAC's SINAD and linearity are 16 bit vs 21 bit? Using the principle - voiced by Amir and others in many contexts and threads - of 10dB of headroom, a DAC that provides only 15 bits' worth of SINAD and linearity is perfectly sufficient to play 13-bit digital content. You can get a DAC that Amir would describe as a broken implementation and still not come anywhere close to reducing the fidelity of an undecoded MQA file. That is, to put it politely, an undesirable scenario for 99% of the members here - and I would say most people would call it simply unacceptable.
• It creates new, mostly hidden costs for equipment. Except for some Topping gear that's available with and without MQA capability, MQA raises the price of equipment because it imposes licensing fees. Whether we pay more for the product or companies absorb the cost is irrelevant. We pay more, or else companies have less to spend on R&D, engineering, build quality, other features, or whatever. If it adds cost and provides - per point #1 above - no sonic benefit, then it's parastic and undesirable.
• It potentially compromises the sonics of everything that goes through an MQA-enabled DAC. For those DACs that keep MQA's digital reconstruction filter permanently engaged, MQA means that all of your music is locked into using that filter. Slow, leaky filters like MQA's are detrimental to the sonics of the audio band because they increase aliasing, which creates needless distortion. Amir consistently points this out - and his recent claim not to want to comment on MQA's filtering makes no sense since he routinely comments quite clearly on MQA-style filtering; he quite rightly condemns it.

Now, for whatever reason, Amir has chosen to take all of these objections to MQA and lump them under the heading of "people who are not in the industry waving their hands and not understanding how the business works and not understanding technology." That is not a persuasive argument, and I'm honestly mystified as to why he is so insistent on making it.

Regardless, the issues mentioned here - and not stupidity or ignorance - are why the MQA discussions persist.

Thank you!

The answer to your first question is yes, largely due to their "deblurring" and lowering the noise in older recordings. My post only dealt with the accusation of LOSSY, which is meaningless, once the loss is below audibility.

The answer to your second question is yes, but for only one of your reasons - slight sonic degradation for people who don't have a decoder AND can hear above 16KHz. The rest of your reasons only become relevant if MQA takes over the world and that's clearly not happening.

 

[ebslo]

What was the source file in #1?

This is of course hypothetical, but let's say the source file is 24-bit/96kHz PCM. Let's also say #2 is competently downsampled from that same source.
I realize I've used Nx48kHz instead of the Nx44.1kHz previously under discussion for this hypothetical example, but hopefully that shouldn't change the answer as long as it's consistent.

 

[amirm]

I kinda agree that the OP lacked some knowledge, and he did go way far into his suspicions (even if they may be partly shared by others) but he nonetheless did something useful. But I can't fathom how you could call the scandalous marketing bullshit full of rhetoric tricks used in the blog post response as just an "explanation".

OP did do something useful. He did more than that actually by clearly showing that MQA is not mathematically lossless so no one should remotely assume that.

The rest of your post I don't understand at all. The blog post is full of technical information which I have been using to shine more light on the topic. It is not remotely "marketing bullshit." Here is an example:

This is clear, technical information that pokes a massive hole in the technical relevance of OP's test. There is noting marketing here. It says the worse case musical information they have found out of encoding millions of tracks shows the spectrum to be well less than OP's test clips. It is authoritative, data based, and shows significant depth of knowledge. This makes your characterization completely wrong and indicative of not reading what you are talking about.