MQA Users & Discussion Thread

[PKay]

10 hours ago, awayward said:

I can sort of understand the angst towards MQA, I may be a simple man but MQA music has not restricted my access to music or cost me more to hear it, 

From what I understand it costs the artist. They aren’t interfering in the process for nothing. 

[Volunteer]

13 hours ago, awayward said:

I can sort of understand the angst towards MQA, I may be a simple man but MQA music has not restricted my access to music or cost me more to hear it,

It has cost more if you are doing the full unfolds. It's somewhat hidden and is in the cost of an mqa DAC purchase. One may well have bought a DAC prior to awareness of the existence of mqa and intended to use it on other PCM formats and then discovered Tidal/mqa along the way, but you paid for it nevertheless. 

 

There are a couple of items worth citing in this conundrum:

1. Topping DACs - they offered the same model DAC in 2 variants. One was mqa and the other non-mqa. The mqa variant was more expensive due to a combination of licence and software costs.

2. Roon. An additional factor is that Roon pays mqa ltd for every "first unfold" of an mqa track that it performs. The remaining processing is done in the DAC - be it an mqa DAC, in which case further mqa processing occurs, or a non-mqa DAC, in which case the track is playable by non-mqa DACs with varying degrees of subjective impacts. This shows there is a financial consideration to mqa ltd involved; it's just subsidised by all users.

 

Roon is not used by many, but it's a useful example in terms of breaking down the economics and costs of mqa to the streaming community as a whole.

 

On streaming costs, your mqa has been subsidised by others who didn't want mqa. In actual fact, mqa has cost many others to incur additional costs in getting to Redbook or Hi-Res versions as they would have to utilise additional services such as Qobuz to do so. Redbooks are being systematically culled by Tidal.

 

13 hours ago, awayward said:

and IMHO some MQA music sounds better than the equivalent non MQA formats

Be careful here. Whilst the reported benefits of mqa are touted as being at their pinnacle when doing the end-to-end decode including secondary unfolds within an mqa certified DAC, there are also other reasons why mqa *might* sound better than PCM or HiRes that are nothing to do with any claimed superiority of mqa as a format.

 

There has been a recommendation from mqa ltd to DAC manufacturers to push all formats, both mqa and non-mqa, through an initial mqa upsampling process. There is an mqa ltd claim that mqa upsampling helps to stop pops/clicks on the DAC when swapping between mqa and non mqa formats.

 

Now what do you suppose mqa upsampling does to non mqa material? Whilst I do not know the explicit answer on this, an enquiring mindset could ask whether (by design or by accident) something deleterious happens to a non mqa stream being processed this way - wouldn't the mqa material then be seen as better, even if artificially? The fact is, that the mqa upsampling process has shown outcomes that are specifically, identifiably and measurably different for PCM tracks (non mqa) that have gone through the process. i.e. the resultant output is different than had it been upsampled on a non mqa DAC or where the manufacturer's implementation had not followed mqa ltd guidance on upsampling (looking at good guys, like Lumin )

 

By way of example (and not using even Redbook here), when I introduced a non mqa DAC into my setup and connected my player to it (thereby avoiding the inbuilt mqa DAC and processes), I noticed a very clear improvement on non mqa material. Not even the Redbook stuff, let alone HiRes. I noticed it on Spotify of all things. What the hell had been going on? Was it mqa hobbling via its upsampling? I don't know. Admittedly my new DAC is a significant uplift in quality and engineering, but this can't be the full answer. Spotify? Really? Of all things... Little, crappy, poor resolution Spotify streams... Suddenly becoming listenable on my mid-table HiFi system. Go figure.

 

There is not enough transparency in the DAC world about who is following the mqa ltd recommendations and who is not with regards to mqa upsampling. But one thing is for sure, if some mqa DACs are producing measurable impacts from this process on PCM or HiRes output streams, then even if you don't use Tidal and avoid mqa sources, you will have to acquire a DAC that is non mqa (or a DAC manufacturer, like Lumin, that demonstrably treats PCM properly without mqa upsampling) in order to hear music unencumbered in anyway by the insidious, creeping and highly questionable nature of this parasitic format.

 

TL;DR - mqa is bad, mmmkay? 

 

Edited August 4, 2021 by El Tel
Punchu.. punktoo.. puntchoowashu. Spelling.

[rantan]

39 minutes ago, El Tel said:

There is not enough transparency in the DAC world about who is following the mqa ltd recommendations and who is not with regards to mqa upsampling. But one thing is for sure, if some mqa DACs are producing measurable impacts from this process on PCM or HiRes output streams, then even if you don't use Tidal and avoid mqa sources, you will have to acquire a DAC that is non mqa (or a DAC manufacturer, like Lumin, that demonstrably treats PCM properly without mqa upsampling) in order to hear music unencumbered in anyway by the insidious, creeping and highly questionable nature of this parasitic format.

 

This is a truly excellent summary of why MQA is a blight on any music format  and we can only hope it dies a quick and painful death.

 

 

 

P.S. My apologies to everybody here for my OT post/s. With 20/20 hindsight it was discordant and not at all helpful.

 

Edit: typos

Edited August 4, 2021 by rantan

[tripitaka]

2 hours ago, El Tel said:

It has cost more if you are doing the full unfolds. It's somewhat hidden and is in the cost of an mqa DAC purchase. One may well have bought a DAC prior to awareness of the existence of mqa and intended to use it on other PCM formats and then discovered Tidal/mqa along the way, but you paid for it nevertheless. 

 

There are a couple of items worth citing in this conundrum:

1. Topping DACs - they offered the same model DAC in 2 variants. One was mqa and the other non-mqa. The mqa variant was more expensive due to a combination of licence and software costs.

2. Roon. An additional factor is that Roon pays mqa ltd for every "first unfold" of an mqa track that it performs. The remaining processing is done in the DAC - be it an mqa DAC, in which case further mqa processing occurs, or a non-mqa DAC, in which case the track is playable by non-mqa DACs with varying degrees of subjective impacts. This shows there is a financial consideration to mqa ltd involved; it's just subsidised by all users.

 

Roon is not used by many, but it's a useful example in terms of breaking down the economics and costs of mqa to the streaming community as a whole.

 

On streaming costs, your mqa has been subsidised by others who didn't want mqa. In actual fact, mqa has cost many others to incur additional costs in getting to Redbook or Hi-Res versions as they would have to utilise additional services such as Qobuz to do so. Redbooks are being systematically culled by Tidal.

 

Be careful here. Whilst the reported benefits of mqa are touted as being at their pinnacle when doing the end-to-end decode including secondary unfolds within an mqa certified DAC, there are also other reasons why mqa *might* sound better than PCM or HiRes that are nothing to do with any claimed superiority of mqa as a format.

 

There has been a recommendation from mqa ltd to DAC manufacturers to push all formats, both mqa and non-mqa, through an initial mqa upsampling process. There is an mqa ltd claim that mqa upsampling helps to stop pops/clicks on the DAC when swapping between mqa and non mqa formats.

 

Now what do you suppose mqa upsampling does to non mqa material? Whilst I do not know the explicit answer on this, an enquiring mindset could ask whether (by design or by accident) something deleterious happens to a non mqa stream being processed this way - wouldn't the mqa material then be seen as better, even if artificially? The fact is, that the mqa upsampling process has shown outcomes that are specifically, identifiably and measurably different for PCM tracks (non mqa) that have gone through the process. i.e. the resultant output is different than had it been upsampled on a non mqa DAC or where the manufacturer's implementation had not followed mqa ltd guidance on upsampling (looking at good guys, like Lumin )

 

By way of example (and not using even Redbook here), when I introduced a non mqa DAC into my setup and connected my player to it (thereby avoiding the inbuilt mqa DAC and processes), I noticed a very clear improvement on non mqa material. Not even the Redbook stuff, let alone HiRes. I noticed it on Spotify of all things. What the hell had been going on? Was it mqa hobbling via its upsampling? I don't know. Admittedly my new DAC is a significant uplift in quality and engineering, but this can't be the full answer. Spotify? Really? Of all things... Little, crappy, poor resolution Spotify streams... Suddenly becoming listenable on my mid-table HiFi system. Go figure.

 

There is not enough transparency in the DAC world about who is following the mqa ltd recommendations and who is not with regards to mqa upsampling. But one thing is for sure, if some mqa DACs are producing measurable impacts from this process on PCM or HiRes output streams, then even if you don't use Tidal and avoid mqa sources, you will have to acquire a DAC that is non mqa (or a DAC manufacturer, like Lumin, that demonstrably treats PCM properly without mqa upsampling) in order to hear music unencumbered in anyway by the insidious, creeping and highly questionable nature of this parasitic format.

 

TL;DR - mqa is bad, mmmkay? 

 

 

IMO this excellent summary demonstrates why it is not right that this thread remains in the Great Audio Debate section (which deals with debates about subjective imponderables).

[Volunteer]

2 minutes ago, tripitaka said:

 

IMO this excellent summary demonstrates why it is not right that this thread remains in the Great Audio Debate section (which deals with debates about subjective imponderables).

I wish more people understood just how much of a threat it is.

 

It's a determined attempt to subvert the integrity of the audio chain end-to-end in order to exploit it financially. If this allowed to become all pervasive, then it's a matter of time before mechanisms are introduced to further hamper the open standards we currently enjoy. Data can be forced through encryption, for example, which might only be decrypted by certified equipment or specialist software including, but not limited to, ongoing additional subscription cost.

 

It's a slippery slope.

[Volunteer]

31 minutes ago, tripitaka said:

....demonstrates why it is not right that this thread remains in the Great Audio Debate section (which deals with debates about subjective imponderables)

 

I appreciated the sentiment of this, but initially had some reservations about whether the thread was appropriate for removing from the subjectivity-pig-pen that is the Great Audio Debate or not. 

 

The more I thought about it though, the more inclined I am to agree. The core topic is not about audio quality or service excellence etc, it is about whether or not the framework of the format and the surrounding issues are in anyway fraudulent.  The format's technical operating mechanisms wholly within its own proprietary chain are not what is in question.

 

The one thing I have often thought is that the only possible benefit of mqa is the idea that new recordings coming from any studio or mastering process could be preserved the whole way through to the consumer’s player. (Despite how it may seem, I am open to changing my opinion on things when presented with new and compelling evidence that is happy to be challenged)

 

However, the practices surrounding it, are most definitely worthy of debate and are less disposed to subjectivity and more likely to have light thrown on the veracity of the claims when we look around the periphery of the format. These ones spring to mind:

• Implementation and impacts (particularly where source material is not mqa) in an otherwise open ecosystem
• Methods and labelling (particularly surrounding older recordings and masters that are batch-processed for user consumption)
• Potential collusion to reduce consumer choice

[PKay]

1 hour ago, tripitaka said:

 

IMO this excellent summary demonstrates why it is not right that this thread remains in the Great Audio Debate section (which deals with debates about subjective imponderables).

I wholeheartedly agree and I initially posted it in the digital section. @Marcmoved it to this section. 

[rantan]

1 minute ago, PKay said:

I wholeheartedly agree and I initially posted it in the digital section. @Marcmoved it to this section. 

 

100%

 

MQA is absolutely, positively, definitely, not designed to benefit the end user in any way.

 

No debate required.

[LHC]

6 hours ago, El Tel said:

Potential collusion to reduce consumer choice

 

There is no need to debate the existence of this 'potential', it certainly exists. What is debatable is whether they have actually realise this potential. If you wish to discuss this aspects then this debate forum is highly appropriate. 

[LHC]

7 hours ago, tripitaka said:

 

IMO this excellent summary demonstrates why it is not right that this thread remains in the Great Audio Debate section (which deals with debates about subjective imponderables).

 

It is debatable whether this section deals only with subjective matters. If you go through other threads there are example of disagreement over objective topics as well. 

[LHC]

9 hours ago, El Tel said:

There are a couple of items worth citing in this conundrum:

1. Topping DACs - they offered the same model DAC in 2 variants. One was mqa and the other non-mqa. The mqa variant was more expensive due to a combination of licence and software costs.

2. Roon. An additional factor is that Roon pays mqa ltd for every "first unfold" of an mqa track that it performs. The remaining processing is done in the DAC - be it an mqa DAC, in which case further mqa processing occurs, or a non-mqa DAC, in which case the track is playable by non-mqa DACs with varying degrees of subjective impacts. This shows there is a financial consideration to mqa ltd involved; it's just subsidised by all users.

 

Roon is not used by many, but it's a useful example in terms of breaking down the economics and costs of mqa to the streaming community as a whole.

 

On streaming costs, your mqa has been subsidised by others who didn't want mqa. In actual fact, mqa has cost many others to incur additional costs in getting to Redbook or Hi-Res versions as they would have to utilise additional services such as Qobuz to do so. Redbooks are being systematically culled by Tidal.

 

The argument against MQA about costs here is not a strong one. The costs mentioned here may not be very transparent or explicit, but they are no more hidden than other normal form of business that we are familiar with in day-to-day transactions. For example in order to take advantage of After-Pay, retailers actually have to pay After-Pay a fee for its service. So this is a cost that retailers could pass on to consumers by rising the prices. Same thing can be said about other add-on services like food delivery (Uber Eat etc). As long as those cost are incurred in ways not in violation of retailing laws, the is no fraud. In a market ecosystem some subsidisation will occur, but that is the nature of how the system works. This certainly can help keep the 'cost of entry' to MQA low enough for those who wish to adapt it. 

[LHC]

10 hours ago, El Tel said:

There has been a recommendation from mqa ltd to DAC manufacturers to push all formats, both mqa and non-mqa, through an initial mqa upsampling process. There is an mqa ltd claim that mqa upsampling helps to stop pops/clicks on the DAC when swapping between mqa and non mqa formats.

 

As discussed previously, we should be careful when discussing this.  This example occurred for one particular DAC manufacturer who made errors in its implementation. This was exposed and criticised by Stereophile's JA. The issue with pops/clicks have been reported by other reviewers so it is a real one. So if MQA's recommendation does address this issue, then there is nothing maleficence; and any negative effects on non-mqa tracks should certainly be fixed. As those negative effects can be measured on a test bench, it in MQA's interest to ensure it won't occur again. 

 

Maybe a longer term issue may be this: let assume MQA can come up with an upsampler that can pass-through non-mqa signal with no measurable negative effects (compare with a comparable non-mqa upsampler). A DAC manufacturer would be tempted to save cost by using the MQA upsampler only, and not create separate pathways for mqa and non-mqa signals.  This cost cutting could certainly happen in budget and mid-price DAC segments - potentially leading to a dominance by MQA hardware there in the longer term. 

[zippi]

All the more reason to favour NOS TDA philips chip DACs and NOS R2R DACs in my book.

 

I cannot stomach the semantic acrobatics in the supposed justifications of this ADC-DAC audio origami that's at the core of MQA. And all these - "no discernible difference arguments". Discernible by whom, when and how? And if that is so - why pursue the HiRes digital lossless audio at all. 320mp3 plays "at no discernible difference" at some level. Of course it doesn't (once it's scrutinised properly) and there are differences. And if not in these early stages (for MQA) it will certainly show up with MQA in the long run - especially in terms of phase response and artefacts as well as the SQ. Small sample sizes currently really skew the SQ in favour of MQA. And even so, many of these issues are showing up and being highlighted all over the place. In the long run with indiscriminate MQA-ing and forcing the upscale-ing DAC paths through MQA algorithms will be a neverending source of bastardisation of the lossless audio signals and unfolding into who knows what.

 

And this is only the tip of the iceberg as the most serious implications in the long run will be with licencing and ownership of the streamed HiRes (or any --- ALL?!) music imprisoned by the MQA origami acrobatics.

 

 

Edited August 4, 2021 by zippi
meant HiRes not HD

[sir sanders zingmore]

2 hours ago, LHC said:

This example occurred for one particular DAC manufacturer who made errors in its implementation.

i expect it was a deliberate choice rather than an error 

 

whilst not as egregious, the Mytek Brooklyn forces you to turn off MQA if you don’t want the filter applied. 
At least it gives the option but it should never force non mqa material through an mqa filter 

Edited August 4, 2021 by sir sanders zingmore

[Steffen]

1 hour ago, zippi said:

All the more reason to favour NOS TDA philips chip DACs and NOS R2R DACs in my book.

 

I don’t think the DAC chips are to blame. The (AKM-based) Topping D90 for example used the same AK4499 DAC chip in the both the MQA and non-MQA versions of the device. The difference was in the XMOS microcontroller (XU208 for non-MQA vs XU216 for MQA).

[LHC]

17 minutes ago, Steffen said:

 

I don’t think the DAC chips are to blame. The (AKM-based) Topping D90 for example used the same AK4499 DAC chip in the both the MQA and non-MQA versions of the device. The difference was in the XMOS microcontroller (XU208 for non-MQA vs XU216 for MQA).

 

The difference between the two is significant according to https://apos.audio/blogs/news/topping-d90-mqa-review-majestic-quick-accurate?_pos=2&_sid=f22b778e3&_ss=r

 

"Another major improvement was utilizing the newest and the highest performance XMOS interface that was developed to date, the XU-216. Simply put, this interface has double the processing power (2000 MIPS) of XU-208 that sits in the regular D90, it has also 4 times the RAM size of second-generation XU-208 and 8 times the RAM capacity of first-generation X-U8. If this reads like some gibberish non-sense, in simpler words D90 MQA has the potential of sounding better than the regular D90 on the USB input, especially with Hi-Res PCM and DSD material that requires more RAM and much higher processing power." 

[zippi]

22 minutes ago, Steffen said:

I don’t think the DAC chips are to blame.

 

I may have been a bit too blunt there - apologies. Not saying DAC chips are to blame however if there's limited or no scope for implementing the MQA at all - as in the cases of these perhaps outdated and non-mainstream chips and DAC architectures (NOS etc) I see it as a plus. I'm not much into DACs anyway as vinyl is my go to source. Do own a DAC nevertheless and I'm thinking any furture purchase of a DAC for me will be one with no MQA support or even any association with MQA at all.

[tripitaka]

17 hours ago, PKay said:

I wholeheartedly agree and I initially posted it in the digital section. @Marcmoved it to this section. 

 

Any chance you could reword the title, to avoid distracting debate about the legal implications of the term 'fraud'

Personally, I always assumed you were using that word in the (non-legal) sense of referring to 'marketing trickery' or 'without substance'.

Cheers

 

[PKay]

16 minutes ago, tripitaka said:

 

Any chance you could reword the title, to avoid distracting debate about the legal implications of the term 'fraud'

Personally, I always assumed you were using that word in the (non-legal) sense of referring to 'marketing trickery' or 'without substance'.

Cheers

 

I copyied the wording from Dr AIX and it is a question so not sure how it could have legal implications?

 

I think people would have less issues if it was only marketing trickery or without substance, but based on a lot of the discussion here and the original video it does have substance and not necessarily positive.  It also removes choice on Tidal - in the early days you could choose MQA or standard  but they took away that option.  I have since and closed my Tidal account, however sadly I can't get Qobuz working on LMS and now I need to wait for Spotify to go "hires".

[tripitaka]

4 hours ago, PKay said:

I copyied the wording from Dr AIX and it is a question so not sure how it could have legal implications?

 

I think people would have less issues if it was only marketing trickery or without substance, but based on a lot of the discussion here and the original video it does have substance and not necessarily positive.  It also removes choice on Tidal - in the early days you could choose MQA or standard  but they took away that option.  I have since and closed my Tidal account, however sadly I can't get Qobuz working on LMS and now I need to wait for Spotify to go "hires".

 

Oh that's a pity about Qobuz for you, hope Spotify gets its act together soon

Re fraud, I wasn't suggesting that Stereonet was exposed, I only meant that the thread is probably located under the Great Audio Debate section because of that particular choice of working, since nobody can demonstrate fraud is occuring here, in the legal sense.

 

Anyway I am hoping that DAC manufacturers are following the issue with enough interest by now...

Edited August 5, 2021 by tripitaka

[POV]

23 hours ago, El Tel said:

I wish more people understood just how much of a threat it is.

 

It's a determined attempt to subvert the integrity of the audio chain end-to-end in order to exploit it financially. If this allowed to become all pervasive, then it's a matter of time before mechanisms are introduced to further hamper the open standards we currently enjoy. Data can be forced through encryption, for example, which might only be decrypted by certified equipment or specialist software including, but not limited to, ongoing additional subscription cost.

 

It's a slippery slope.

 

I think many of us that use Tidal get it.  The main issue for me though is that I still find Tidal to be a far superior platform to Qobuz in terms of user interface, and library content.  My hope is that Spotify Hifi lands in Australia soon and is on par for sound quality with Tidal.  This would enable many users to move away from Tidal without Qobuz being the only real alternative.  Of course for those of us that use Roon there would be a great loss in moving away from having your stream service integrated.  Whilst I would love it if Spotify Hifi gets integrated into Roon, it seems doubtful given the tiny footprint that Roon has compared to the massive international user community Spotify has...

 

So do I take the big hit to my music listening experience on moral grounds or not.    Let's be honest, this is not the only moral dilemma we face in daily life!  Time will tell, but I'm holding out hope that Spotify holds the key...

[Volunteer]

2 hours ago, POV said:

 

I think many of us that use Tidal get it.  The main issue for me though is that I still find Tidal to be a far superior platform to Qobuz in terms of user interface, and library content.  My hope is that Spotify Hifi lands in Australia soon and is on par for sound quality with Tidal.  This would enable many users to move away from Tidal without Qobuz being the only real alternative.  Of course for those of us that use Roon there would be a great loss in moving away from having your stream service integrated.  Whilst I would love it if Spotify Hifi gets integrated into Roon, it seems doubtful given the tiny footprint that Roon has compared to the massive international user community Spotify has...

 

So do I take the big hit to my music listening experience on moral grounds or not.    Let's be honest, this is not the only moral dilemma we face in daily life!  Time will tell, but I'm holding out hope that Spotify holds the key...

For sure. Everyone makes their own choices. It's an ethical and convenience balance. For example, some people are not buying goods made in China for either human rights concerns or due to trade tensions or for reasons related to protectionism etc. You gotta do what you gotta do. Personally, I accept the short-comings in the Qobuz catalogue and lean on vinyl and Spotify too. Even Spotify sounds good with my non mqa DAC.

 

I can't see Roon ever integrating with Spotify as Roon need far more than an API to make it work with their impressive algorithm, suggestions and rich data viewer. Same for Apple. 

[LHC]

If audiophiles wish to use their personal ethics and values for purchasing decisions, then shouldn't that approach be applied consistently? As posted previously Apple Music have been charged for antitrust and anticompetitive practices. This is far more serious than the MQA situation (MQA have not been charged for anything yet). Shouldn't audiophiles boycott Apple Music and their services/products in protest? Where is the outrage over Apple?

 

 

Edited August 5, 2021 by LHC

[tripitaka]

11 minutes ago, LHC said:

If audiophiles wish to use their personal ethics and values for purchasing decisions, then shouldn't that approach be applied consistently? As posted previously Apple Music have been charged for antitrust and anticompetitive practices. This is far more serious than the MQA situation (MQA have not been charged for anything yet). Shouldn't audiophiles boycott Apple Music and their products in protest? Where is the outrage over Apple?

 

 

I've been outraged by Apple since forever and have uniformly boycotted all their products... but you are correct that one does get exhausted by trying to boycott everything that is imperfect.

 

Still, I see music as such a fundamental, non-negotiable human right, that I am simply angry at even the possibility of a company exercising lingering control over music that I have already paid the artist for. You might say I am 'triggered' and may not always respond rationally

 

 

 

 

Edited August 5, 2021 by tripitaka

[LHC]

22 hours ago, sir sanders zingmore said:

i expect it was a deliberate choice rather than an error 

Only the manufacturers know for sure. 

 

One thing we do know is that Stereophile and JA was at least transparent here - they wrote about the recommendation from MQA and not censored that piece of information.

 

22 hours ago, sir sanders zingmore said:

whilst not as egregious, the Mytek Brooklyn forces you to turn off MQA if you don’t want the filter applied. 
At least it gives the option but it should never force non mqa material through an mqa filter 

 

I agree consumers should be given the maximum of choices. It is also much preferable if non-MQA material are not routed through any MQA associated components. This may be hard to avoid if MQA does become ubiquitous and the cost of entry is exceedingly low. 

 

Another challenge is that it may be difficult to evaluate performance with/without MQA. As the Topping D90 DAC example showed, when they introduced the MQA enabled unit they also upgraded their microprocessor. So it is no longer an apple-to-apple comparison to the non-MQA unit.