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

[levimax]

Not at all. You can send the digital bitstream however you like to an MQA DAC and it will decode and play it. Roon is not needed.

Sorry I left out EQ as that was what I was responding to. As far as I know if you want both EQ and "final unfold" the only solution is Roon and a MQA DAC. (unless you re-digitize the output of your MQA DAC with an ADC and run it though EQ and then out to another DAC but that is not a great solution I don't think.)

 

[BinkieHuckerback]

Question has nothing to do with Tidal. It was about mastering MQA CDs. Physical media at mandated 16 bits, not the normal 24-bit files MQA produces for online distribution in Tidal.

As to Neil Young, I don't care what he said. He didn't conduct any controlled test to show that his content had degraded, or even changed.

...Neil Young drove some celebrities around in his car, I'll have you know...

 

[Raindog123]

That link doesn't open. And the fact that it is 20 to 20K answers your question. It has no ultrasonics so MQA can just pass it through with zero trouble.

Sorry, fat fingers, try now:
https://tidal.com/track/29581922

And, it was a point, not a question… Of a constant-power-level test signal. Well ‘ultrasonic’ to my ears, BTW. And look who is obsessing with MQA now.

 

[symphara]

I am not sure which streamer does the "full unfold" or not but the problem is EQ. You can't EQ before you send information to your MQA DAC because it changes the "bits" and the light won't come on and you won't get the "final unfold". Roon cut a deal with MQA where you can apply EQ to the MQA stream and then Roon "reauthorizes" the stream so the MQA light comes on and you get the "final unfold". If you don't care about the "final unfold" then it doesn't matter.

I cannot EQ because my DAC pulls music directly from Tidal, and my DAC has no EQ, so there‘s no opportunity to change the “bits“ for EQ. I’m not aware of any reason as to why a DAC couldn’t implement EQ similarly to Roon. I don’t have Roon. I have no idea why EQ couldn’t be implemented post-MQA decoding. Nor am I bothered about not having EQ.

The fact is that very few DACs actually have any EQ whatsoever. I can only think of miniDSP gear.

PS: my DAC doesn’t have the infamous “light”. It just displays on the screen (and in the app) if it decodes MQA.

 

[Dennis_FL]

??? 95% of what I play on Tidal is NOT MQA. MQA makes up a tiny fraction of content I find on Tidal. Most of the time I have to seek it out to find it. Please don't run with folks who have other reasons to hate MQA. Chris who wrote that is making a business out of anti-MQA. He is not technical and doesn't understand any of these topics himself.

Maybe I was a bit harsh. I’m going to find a title that I have in DSD and MQA (my Gustard X16 plays both) and see if I can set up a blind test with headphones and speakers. I do remember the first time I heard an MQA it sounded pretty good.

 

[tmtomh]

??? You sell advertising based on the number of views you have on your site. The more food fights, the more views you get and the higher you can charge the advertiser. And the higher your ranking against other competing audio sites.

I suggest you watch his video he did with Jason on his outlook on life there. When you make a living from a site and it is what pays your bills, this is what you do. You think every minute about how you are going to improve your ranking/traffic.

I ran another forum with someone else who all of a sudden wanted to make money. I told him the price: that you have to allow food fights and whether he was OK with it. He said yes. Food fights came and then he changed his mind and said let's stop the food fight and get rid of all the objectivist talk. I told him that I had warned him about this and that he couldn't have his cake and eat it too.

Really, none of this matters in the larger context. Chris has no business taking up religion around objectivity in MQA. If he cares about MQA one way or the other, he should do his subjectivist listening tests and publish that. Same with OP. OP says Schiit Magnius headphone amp lowers dynamic range by increasing the level of low level sounds. If he is so good at determining this by ear, then he should have performed a bunch of listening tests with MQA with real music and educate his audience that way. Don't go running test tones though a music codec and pretend with straight face you care. He doesn't and neither does Chris.

Neither I nor anyone else is disputing that Chris makes his money from views on AS (it's one of the reasons I stopped participating after his objectivist purge - I didn't want to contribute to his page views after that). But you made a specific charge that he's become anti-MQA in order to generate those views, and solely for that reason. There is plenty of empirical evidence that undermines that claim, including (a) 99% of the massive MQA thread over there all happened before Chris became vocally anti-MQA; and (b) Chris has worked to put out many fires and endured many headaches around other dust-ups there, which he would not have bothered to try to manage had he just wanted to generate page-views by facilitating flame wars.

I agree with you that "none of this" - "this" being whether he's anti-MQA because he sees money in that - "matters in the larger context." You're the one who brought it up.

I certainly agree with you that Chris is ill-equipped to get into detail-oriented battles about technical matters, given his lack of expertise, his subjectivist commitments, and his (to my eyes at least) not-terribly-rigorous way of thinking about and writing about technical matters.

 

[earlevel]

What is the criteria for design of this speaker and a million like it?

Isn't that the high frequencies are played at much lower levels and if you push them overwise, you will damage it?

Statistical solutions are very helpful in designing workable systems. Trying to make a solution 100% massively increases cost which is the problem with FLAC encoding high sample rate/high-res content. I have tracks that are nearly one gigabyte in size for just one song!

View attachment 133186

This is the cost of "encode anything."

No lossy encoder would exist ever if we applied the same standard. Corner cases can be terrible sounding. But to the extent that 99.999% of content does well, the job is done. If you want that last 0.001% then, the solution is not for you can seek out options like above.

Look, I'm not new to this game. I've run unfiltered white noise at ear-splitting volume into my sound system many, many times. Geez, of course diving for the fader or patch cord or whatever to make it stop. Mistake are often made in modular synths, or software. I've never blown a tweeter that way. OTOH, a split second on a clipped signal (was developing a guitar amp simulator), sure—fried a midrange driver that way.

But you're acting like white noise will destroy peoples' speakers. Sure, at enough volume for enough time, depending on the speaker—no speaker will melt down with white noise at reasonable listening levels. But you can't tell me what MQA barfs on, you say 0.001% but you don't know that pertains to MQA, neither of us has any idea. Bob says MQA will put out "nonsense" with non-musical signals (presumably, it fails to encode, not actually make a nonsense audio file). Surely there is some objective way to say at what point MQA fails.

With your FLAC failure example, we can always choose to not compress, and not lose one bit of fidelity with a non-compressed version that plays on the same decoder. That would not be true of MQA, so it's a much bigger deal.

 

[ebslo]

I’m flabbergasted that MQA is called “broken” (or worse) in this thread.

It's a matter of context. If MQA is marketed as "lossless" then it is fair to evaluate it as such; since it obviously fails these tests, the term "broken" is accurate within that context. They have since backed off from that claim somewhat, at least to the point of plausible deniability, so I will, going forward, hold it to standards appropriate for a lossy codec.

 

[tmtomh]

What is the criteria for design of this speaker and a million like it?

The pertinent point is that the criteria for design of this speaker are different than the criteria for design of an amp, which are different than the criteria for design of a DAC, which are different than the criteria for design of the digital formats for the music played through all this equipment - as you well know.

 

[levimax]

I cannot EQ because my DAC pulls music directly from Tidal, and my DAC has no EQ, so there‘s no opportunity to change the “bits“ for EQ. I’m not aware of any reason as to why a DAC couldn’t implement EQ similarly to Roon. I don’t have Roon. I have no idea why EQ couldn’t be implemented post-MQA decoding. Nor am I bothered about not having EQ.

The fact is that very few DACs actually have any EQ whatsoever. I can only think of miniDSP gear.

PS: my DAC doesn’t have the infamous “light”. It just displays on the screen (and in the app) if it decodes MQA.

No EQ no worries. Again the problem with EQ is that an MQA DAC looks for make sure the stream has not been "altered" before the light comes on and the final unfold happens. Any EQ will change the stream and the DAC won't recognize the stream as MQA. Roon changes the stream with EQ and then "reauthorizes" it as genuine MQA and sends it to the DAC. You need an agreement with MQA to do this.

Of course you could EQ after the MQA DAC but most people like to do that in the digital domain which would mean MQA DAC > ADC > EQ Processing Device>Another DAC . Or you could use an analog equalizer but most on these boards think that adds too much noise and distortion in addition to being expensive for a good one.

The bottom line is MQA is not EQ friendly without Roon although that can change

 

[DimitryZ]

You are correct… It is common, as soon as it does not violate Shannon Limit - of only that much information fitting in that much BW…

But that’s not the point. My point is that such band-splitting of a PCM waveform, when based on the frequency band/bin dynamic range - while doable - is definitely beyond your average ‘let me draw you a triangle’ hand-waiving.

It seems that all professionals that took a look at it, like Jim Lesurf, had no problems understanding it.

For the consumer, an understanding that MQA bandsplits and artefacts from this operation is inaudible should be enough.

 

[DimitryZ]

Look, I'm not new to this game. I've run unfiltered white noise at ear-splitting volume into my sound system many, many times. Geez, of course diving for the fader or patch cord or whatever to make it stop. Mistake are often made in modular synths, or software. I've never blow a tweeter that way. OTOH, a split second on clipped signal (was developing a guitar amp simulator), sure—fried a midrange driver that way.

But you're acting like white noise will destroy peoples' speakers. Sure, at enough volume for enough time, depending on the speaker—no speaker will melt down with white noise at reasonable listening levels. But you can't tell me what MQA barfs on, you say 0.001% but you don't know that pertains to MQA, neither of us has any idea. Bob says MQA will put out "nonsense" with non-musical signals (presumably, it fails to encode, not actually make a nonsense audio file). Surely there is some objective way to say at what point MQA fails.

With your FLAC failure example, we can always choose to not compress, and not lose one bit of fidelity with a non-compressed version that plays on the same decoder. That would not be true of MQA, so it's a much bigger deal.

You obviously never owned ribbon Magnepans. They have a special program for affordable tweeter exchange.

 

[7ryder]

I find that surprising myself since this is your 2nd post here, 1st one was making some menial point to Amir on behalf of Goldenone 4.5 weeks ago... unless you're ringing in from PFM?



JSmith

I've never been on PFM. My comment was based on how rude @DimitryZ has been at times since first posting on 5/29.

While I'm not a fan of MQA, he does appear to know his stuff, but if this was his style at PFM, I can understand what led to him being banned.

 

[symphara]

It's a matter of context. If MQA is marketed as "lossless" then it is fair to evaluate it as such; since it obviously fails these tests, the term "broken" is accurate within that context. They have since backed off from that claim somewhat, at least to the point of plausible deniability, so I will, going forward, hold it to standards appropriate for a lossy codec.

The term “broken” really doesn’t apply. It’s like saying that my cars are “broken” because the Napa leather doesn’t actually come from Napa, California. They work just fine and the leather interior is beautiful.

And it depends on what you consider “losless”. Strictly speaking, no audio reproduction chain is lossless, from the original analog sound, no matter what equipment, media or codecs we use, we just cannot do it. CD isn‘t lossless either.

The container Tidal uses is lossless (in computing terms) so there’s that. I’m not that bothered by lofty marketing claims. While I kindof understand your point, I don’t get the hatred and fervor against MQA (worthy of better goals) on this particular subject.

The bottom line is MQA is not EQ friendly without Roon although that can change

Understood. I see this as a DAC failure, and not particularly as an MQA failure. As I said before, MQA is a niche format, and few DACs implement EQ at all, so there’s no market pressure to have DACs with MQA and EQ. There’s no market pressure to have DACs with EQ to begin with.

 

[adamd]

I've never been on PFM. My comment was based on how rude @DimitryZ has been at times since first posting on 5/29.

While I'm not a fan of MQA, he does appear to know his stuff,

It is impossible to overstate how misleading that appearance is

but if this was his style at PFM, I can understand what led to him being banned.

but you are on the money here. (Except that putting the two points together, the style is even worse)

 

[Maki]

What is the criteria for design of this speaker and a million like it?

Isn't that the high frequencies are played at much lower levels and if you push them overwise, you will damage it?

Statistical solutions are very helpful in designing workable systems. Trying to make a solution 100% massively increases cost which is the problem with FLAC encoding high sample rate/high-res content. I have tracks that are nearly one gigabyte in size for just one song!

View attachment 133186

This is the cost of "encode anything."

No lossy encoder would exist ever if we applied the same standard. Corner cases can be terrible sounding. But to the extent that 99.999% of content does well, the job is done. If you want that last 0.001% then, the solution is not for you can seek out options like above.

That's a wav file though and also dsd (?)

 

[earlevel]

Isn't that the high frequencies are played at much lower levels and if you push them overwise, you will damage it?

There are limits, of course—there is a limit at which you can damage a given speaker with a given amp with normal music.

But chirps don't follow "high frequencies are played at much lower levels". I've helped develope speaker/room correction software for a company that makes studio monitors. And REW does this too, and I'm pretty sure has an option for white noise as well. It's all about volume, of course—that Dreamcrusher tune I posted will likely cause damage at power levels and speakers that people own (but your ear will tell you to not play it loud enough to do that). But you can run a chirp or white noise pretty loud without frying speakers.

Just saying, you seem to imply that equal energy per frequency is a menace. It's rare in musical balance, and of course not a full volume, but I have to push back on it being inherently damaging. But I think this discussion on speakers is a tangent, though, because whatever the signal, we tend to avoid damaging levels, and music rarely tricks us with a soft passage to set our expectations, then full blast high frequencies. (But even white noise isn't "full blast high frequencies", as there is a lot of energy taken up with lower frequencies.) The real issue is MQA-encoded audio. In all the messages I've replied to, and our ensuing discussion, you haven't said something like, "MQA can't encode white noise that is at unusually high level, say, -3 dB full scale". You've said that white noise (at least with 44.1k bandwidth) isn't music and should not be expected to encode. I just disagree. I don't expect to change you mind, I'm just registering a voice that disagrees with that point, in this thread.

Anyway, I'm not going to play white noise at volumes to see at what point I damage my speakers or ears, but I've heard it pretty loud in here and and elsewhere and have yet to replace anything. <shrug>

 

[ebslo]

Statistical solutions are very helpful in designing workable systems. Trying to make a solution 100% massively increases cost which is the problem with FLAC encoding high sample rate/high-res content. I have tracks that are nearly one gigabyte in size for just one song!

View attachment 133186

If the size of that file concerns you, may I humbly suggest utilizing the FLAC encoding that you mentioned?

Edit: Nevermind, I see it was already mentioned by @Maki .

 

[symphara]

There are limits, of course—there is a limit at which you can damage a given speaker with a given amp with normal music.

But chirps don't follow "high frequencies are played at much lower levels". I've helped develope speaker/room correction software for a company that makes studio monitors. And REW does this too, and I'm pretty sure has an option for white noise as well. It's all about volume, of course—that Dreamcrusher tune I posted will likely cause damage at power levels and speakers that people own (but your ear will tell you to not play it loud enough to do that). But you can run a chirp or white noise pretty loud without frying speakers.

Just saying, you seem to imply that equal energy per frequency is a menace. It's rare in musical balance, and of course not a full volume, but I have to push back on it being inherently damaging. But I think this discussion on speakers is a tangent, though, because whatever the signal, we tend to avoid damaging levels, and music rarely tricks us with a soft passage to set our expectations, then full blast high frequencies. (But even white noise isn't "full blast high frequencies", as there is a lot of energy taken up with lower frequencies.) The real issue is MQA-encoded audio. In all the messages I've replied to, and our ensuing discussion, you haven't said something like, "MQA can't encode white noise that is at unusually high level, say, -3 dB full scale". You've said that white noise (at least with 44.1k bandwidth) isn't music and should not be expected to encode. I just disagree. I don't expect to change you mind, I'm just registering a voice that disagrees with that point, in this thread.

Anyway, I'm not going to play white noise at volumes to see at what point I damage my speakers or ears, but I've heard it pretty loud in here and and elsewhere and have yet to replace anything. <shrug>

Why should we care about a codec encoding white noise? And sho says that MQA cannot encode white noise? My impression was that MQA can encode white noise, should you wish it for some reason, just perhaps not the MQA encoder used by the OP which was specifically meant for a much narrower audio domain, i.e. music and not audio test patterns. If you wanted to encode white noise with MQA, my understanding is that you have to get a much more sophisticated encoder, which costs money, and the OP didn’t do that.

 

[Raindog123]

Anyway, I'm not going to play white noise at volumes to see at what point I damage my speakers or ears

…and in case you’ll change you mind, they have it on Tidal too (‘White Noise 0dB”):
https://tidal.com/track/29582006