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

[Raindog123]

sn't a simplified summary (for the first unfold) that the encoder is a band splitter, and the decoder joins the bands?

Most definitely. But tying the concept to this "reduced dynamic-range envelope ("triangle") is where I am curious - of 'exactly how?' and 'does it work?' (ie, algorithm and implementation)

 

[tmtomh]

Can you sketch your levels over my cartoon? I'll pretty them up an update (if we'll all agree)

Sure! Although I'm not sure what my sketch can really add - and apologies if I'm misunderstanding what you're asking me to sketch. This image circles your diagram's label showing that the undithered signal covers 13 bits/80dB of range. It also draws arrow lines and labels the peak level and noise floor of that undithered signal: approx. 93dB peak, approx 13dB floor = 80dB as the label says.

Not to be pedantic, but 13 bits isn't actually quite 80dB. To my knowledge it comes out to 78.26dB (6.02dB per bit, AFAIK - happy to be corrected if I'm wrong). Conversely, my prior comment that this is 16dB less than 16-bit redbook was also slightly off: it's really 18.06dB less (3x6.02)

At any rate, for everyone else reading, I think it's worth re-stating and re-clarifying that your diagram, per your earlier comment above, is meant to represent what happens with an "MQA CD," in other words an MQA file in a container format that only has 16 bits' worth of depth, yes?

In that event, and if your diagram is correct, the 3 bottom bits are not simply "dither," because they cannot contain the randomized and/or noise-shaped info dither usually contains - some portion of those 3 bits are needed to contain MQA flags and codec information.

What I'm not clear on is whether or not, in this example, those 3 bits also contain the lossy-encoded ultrasonics we can see in the diagram: There's a bump-out in the red box, from 22.05k to 44.1k (because this would be an 88.2k MQA file not a 96k one), and that bump-out includes ultrasonics only up to about 55-60dB below peak.

My understanding has been that when it comes to MQA CDs or any other MQA file in a 16-bit container, those ultrasonics are simply not encoded - they are discarded because there is nowhere to put them. My understanding has been that only a 24-bit MQA file can accommodate those ultrasonics. That in turn makes MQA CDs completely useless as they sacrifice the bottom 3 bits only for the purposes of containing a flag telling the DAC to use a particular reconstruction filter.

On the other hand, if I am mistaken and MQA CDs/16-bit MQA files put some or all of those ultrasonics in those lower 3 bits, then that makes it even worse because it only adds to the non-random quality of the noise in those lower 3 bits, further degrading the effective noise floor/bit depth.

Regardless, this diagram is a nice illustration of MQA's underlying conviction: MQA is based on the claim that it is worth reducing the unmolested bit depth by 3 bits/18dB in order to preserve low-level ultrasonics. Personally I don't see that as a worthwhile tradeoff.

 

[markus]

MQA is not "lossless" under the standard definition for any high-resolution digital source material.

I know, it's "better than lossless"

 

[DimitryZ]

That might work for you, but I help out with audio equipment recommendations almost everyday over on Reddit. And Tidal or MQA is often part of the conversation. Sometimes several times a week.

Now if you can persuade all of those people to ignore MQA's advertising, that sounds great. But I'm pretty sure you are not capable of that.

I am sorry, that's a tough place to be.

You can go over the @Archimago assessment and maybe adopt his language? Works and sometimes sounds better than FLAC to some listeners, but advertising is over the top...

 

[bennetng]

For anyone who wants to take a look at the 2L-053 MQA files, @mansr generously has made them available to us. Again, the rendered files were made using some reverse engineering of a Bluesound renderer, so no ADC. (assuming I don't misunderstand)

https://drive.google.com/drive/folders/1zc5I2VJnVwBOqweOh8napqABQadnmCoe

Thanks @mansr
Nothing to unfold beyond 40kHz apart from imaging distortion, what's the point.

 

[tmtomh]

MQA has nothing to do with this, don't attempt to hijack my comment. In the case of "crazy audiophiles", MQA can't fulfill their needs either. They would rather keep all the noise in the original DXD files, or filter them with their own methods, for example, with DSP utilties in software players.

Hijacking comments like that appears to be a major part of his M.O., alas.

As to the screenshot you posted just above, yes - there's an absurdity to the preservation of ultrasonics in any format, including but not limited to MQA. In this regard, I think one way of understanding @amirm ’s position on MQA is, if you are going to have a marketplace where there is a demand for high sample-rate digital music content, then MQA's approach is arguably no more silly or flawed than 176.4k and higher PCM, or than DSD - and MQA requires a lot less bandwidth/storage space than 24-bit PCM at sample rates of 176.4 and up.

Of course, that doesn't address the bit-depth problem of MQA. I'm still not quite getting why anyone would think that reduced bit depth would be a better price to pay for ultrasonics than increased file size. I mean, who are the consumers who are sufficiently dedicated to audio to want high-res, and have the disposable income to buy high-res files and/or pay the top Tidal subscription tier price, and yet lack the resources to buy the hard drives and/or internet bandwidth necessary to store and stream 24/96 or 24/192 FLAC?

 

[Raindog123]

Thanks. In other words, any type of content in the audible band should encode fine (down to -100dB or more for MQA, less for MQA-CD, dynamic range extended by dithering similar to what redbook can do). The only signals that will throw off the encoder will be unusual ultrasonics that don't fit in the small ultrasonics box.

Yep. Just need a test-based proof that it works as advertised... Once I was [rather handsomely] paid to build a system that phones to Mars (from Mars, actually)... and instead my system could only call my long-passed grandma... Quite an embarrassment, and a lot of taxpayers money...

 

[tmtomh]

Yep. Just need a test-based proof that it works as advertised... Once I was [rather handsomely] paid to build a system that phones to Mars (from Mars, actually)... and instead my system could only call my long-passed grandma... Quite an embarrassment, actually, and a lot of taxpayers money...

Great story!

RE MQA's encoding, we do need test-based proof. I would think we'd need to see how much of those lower 3 bits is really available for dithering and what the actual noise floor is with a wide variety of music files so that an average and/or a practical limit can be established.

 

[pjug]

Thanks @mansr
Nothing to unfold beyond 40kHz apart from imaging distortion, what's the point.
View attachment 133410

Yes we know anything in the 2nd unfold is not real, but how does the rendering affect the overall fidelity. The spectra of the DXD vs rendered begin to diverge below 20KHz. We already knew that too but at least now we have a track with MQA rendering for any kind of analysis of this.

 

[DimitryZ]

MQA has nothing to do with this, don't attempt to hijack my comment. In the case of "crazy audiophiles", MQA can't fulfill their needs either. They would rather keep all the noise in the original DXD files, or filter them with their own methods, for example, with DSP utilties in software players.

A reply is not a hijack.

The first thing MQA does with a 352 DXD, is to find an ultrasonic music limit and filter above that.

 

[Raindog123]

The practical limitation is ultrasonic level. And I think everyone would agree that outside of man made noise shaped files, actual natural ultrasonic content is very low.

I hope everyone agrees that if it cant be heard, we should not spend our effort (and ultimately customers' money) on it. Eg, find my insinuation about the [missile defense radar] sensors on the forum.

 

[DimitryZ]

Thanks @mansr
Nothing to unfold beyond 40kHz apart from imaging distortion, what's the point.
View attachment 133410

Sure, there is no musical content above that. Makes sense to me.

 

[DimitryZ]

I hope everyone agrees that if it cant be heard, we should not spend our effort (and ultimately customers' money on it). Eg, find my insinuation about the [missile defense radar] sensors on the forum.

Sorry missed that insinuation.

 

[levimax]

. That in turn makes MQA CDs completely useless as they sacrifice the bottom 3 bits only for the purposes of containing a flag telling the DAC to use a particular reconstruction filter.

Don't forget about the DRM like/lite and Watermark content which is stored in those 3 -bits. This point keeps getting made but while the MQA Codec is interesting and fun to try to reverse engineer with limited information it has absolutely no utility to the consumer as anything it can do could be done easier and cheaper with PCM/FLAC (including adding Voodoo DSP / Filtering to make it "sound better than lossless").

The real customers are the record labels and streaming services and what is contained in the 3 bits besides compressed HF information is one of the big selling points. Originally the data saving thing was also a selling point to the streaming services but unfortunately for MQA there is not much utility for the streamers with that anymore.

 

[danadam]

Why not compare it to DXD? It's an increasingly popular format, with lots of content available.

(adding to @tmtomh answer)
Because if MQA is a suitable replacement for DXD in terms of data it keeps, then 24/96 FLAC is also a suitable replacement. So then keeping comparing MQA to DXD and crying how FLAC is inefficient is just disingenuous.

BTW, good to know that you see my messages. I asked twice already about the formula you used:

"24/0.2", where did you get this formula from? AFAIK the null between -24.0 dBFS and -24.2 dBFS signals is at -56.86 dBFS.

If you don't want to answer just say so, I'll understand.

Unless I misinterpreted, did you not imply earlier that comparison even with 20/96 FLAC may be apropos?

Yes, that's why I used word "pretense"

But if we are comparing file size vs. "musical" content (as defined by the literature on behalf of MQA), then it appears comparison between 24-bit/48kHz MQA and 20-bit/96kHz FLAC encoded PCM may indeed be valid, as the extra four low-order bits in the MQA file are being used to store extended-frequency content instead of extra sub-nyquist resolution.

That's very similar to what @voodooless said earlier:

But back to the 20-bit issue. I don't really think it will matter a lot if you feed the MQA decoder a 20-bit or 24-bit file. It will use the lower bits (of the 24-bit output) for the ultrasonic content, so it would chop off some of the 24 bits anyway, it should just chop off less for the 20-bit file. The main point is the file size here. It really makes the whole MQA ordeal unnecessary if they end up being so close. So this definitely is something to consider adding to the bunch of test files for testing.

 

[tmtomh]

The first thing MQA does with a 352 DXD is not "find an ultrasonic music limit and filter above that."

Rather, the first thing MQA does with a 352.8kHZ PCM file is to downsample it to 88.2kHzm, throwing away 75% of the samples in the original file. All MQA encoding then takes place on the downsampled 88.2kHz version.

This is important, because when such a file is played through an MQA renderer (hardware or a hardware-software combo capable of doing full MQA decoding), and it displays at "352.kHz," it is not in any way reconstructing or reproducing the 352.8kHz sample rate of the original. It's just 4x oversampling the 88.2kHz file that the MQA encoder encoded. There really is no such thing as a 352.8kHz (or 384k, or 192k, or 176.4k) original when it comes to MQA, because the encoder can only encode 88.2k or 96k max.

So MQA doesn't "find" anything - it just throws out 3/4 of the samples of a 352.8k file before it does any encoding or filtering. This is - once again - why the comparison of MQA to DXD aka 352.8k PCM only furthers the misleading nature of MQA's communications - and those communications are not merely "advertising" or "marketing materials." Bob Stuart himself has stated, in video interviews, "we put the sample rate back to what it was" - a carefully-chosen phrasing that is very difficult to interpret as anything other than an attempt to cover over what MQA is actually doing (and not doing) with regard to the displayed sample rate of an "authenticated" playback.

 

[DimitryZ]

The first thing MQA does with a 352 DXD is not "find an ultrasonic music limit and filter above that."

Rather, the first thing MQA does with a 352.8kHZ PCM file is to downsample it to 88.2kHz. All MQA encoding then takes place on the downsampled 88.2kHz version.

This is important, because when such a file is played through an MQA renderer (hardware or a hardware-software combo capable of doing full MQAS decoding), and is displays at "352.kHz," it is not in any way reconstructing or reproducing the 352.8kHz sample rate of the original. It's just 4x oversampling the 88.2kHz file that the MQA encoder encoded. There really is no such thing as a 352.8kHz (or 384k, or 192k, or 176.4k) original when it comes to MQA, because the encoder can only encode 88.2k or 96k max.

So MQA doesn't "find" anything - it just throws out 3/4 of the samples of a 352.8k file before it does any encoding or filtering.

Correct, and thank you for pointing that out.

And completely justified in doing that since there is no musically related content in what it threw out.

And I agree with you in that the final upsampling is kind of fakey. And many users stay with a software unfold and don't bother with it.

Can use a non-MQA DAC, too.

 

[Raindog123]

@tmtomh Thanks (for your entire write-up, actually)

In that event, and if your diagram is correct, the 3 bottom bits are not simply "dither," because they cannot contain the randomized and/or noise-shaped info dither usually contains - some portion of those 3 bits are needed to contain MQA flags and codec information.

All, can I ask (everyone) a question "I always wanted and always hesitated to ask":

Can the 'dither' (the 3 bits' in our cartoon) be added by the decoder _after_ the 'ultrasonic information' bits are stripped away? Or does it carry some information (shaping?) all the way from the encoder?

 

[tmtomh]

Correct.

And completely justified in doing that since there is no musically related content in what it threw out.

And I agree with you in that the final upsampling is kind of fakey. And many users stay with a software unfold and don't bother with it.

Can use a non-MQA DAC, too.

• When you write, "Correct" and fail to note that my being correct also means that your prior comment was incorrect, it reads like gaslighting.
• When you write that MQA is "justified" in downsampling 352.8k content to 88.2k, but fail to note that this fact means your repeated attempts to drive the conversation into a comparison of MQA specifically with 352.8k DXD are flawed and misleading, you engage in a method of discussion that is IMHO misleading.
• And when you fail to note that MQA does not simply throw out some ultrasonics, but also compromises bit-depth by retaining other ultrasonics that are just as unnecessary as the ultrasonics it threw out, you are once again engaging in selective discussion that is also IMHO misleading.

I do appreciate your acknowledgment that the final upsampling is “kind of fakey.” Not quite how I’d characterize it, but close enough.

 

[firedog]

Highresaudio is a German online store where you can buy, among other things, MQA files for download. I found the following text on their FAQ which I found interesting.

"First of all, MQA is a lossy codec! The MQA albums that we currently offer are all „MQA-Authenticated“. That means, that we know the origin of the studio master and all files have passed our strict and professional quality control process in order to guaranteed YOU genuine and native Studio Masters.

These MQA albums will remain in our online store. All others that we can’t check and verify its source, we will not offer.

As soon as we have an MQA encoder and quality control software to analysis the MQA encodes, we will offer MQA again. This is something that we are very peculiar and exceptional about - in your interest. Especially if you pay hard earned money.

We are in a very sensible and delicate niche music market. Over the past seven years we have established a very good market position, created a new business for the music industry and artists and customers that cherish the best audible sound reproduction. We moved the music and HiFi-industry into a new business domain, with very little support from anyone. Our USP is that we guarantee (and this is not just said and done) your customers, nothing but the true, native and original source. We can analysis and verify any other audio codec (with MusicScope even DSD and DXD). For MQA is nothing available to assure that the customer is getting our „promise“. We are in the first and front row, selling music and technology to a new and established customer, that truly expects nothing but the real thing!"

They wrote that several years ago. It hasn't been updated and AFAIK, isn't true. They sell the same MQA files everyone else does, I don't think any special QC is going on and I don't think they have anyway of knowing what the master is. The whole "authenticated" part of MQA is BS. All that's authenticated is that they used A master. The quality is for you to accept on faith.