Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[dtc]

BTW, it's worth noting that JRiver has always added playback of any requested file formats, so no worries there.

If this is about MQA, I think it is quite possible that it will not be supported anytime soon unless it is a huge success. That is based on previous comments from Jim and Hendrik.

[JimH]

I don't see the point of MQA.  It seems like a step backward. 

[RD James]

I don't see the point of MQA.  It seems like a step backward. 

The point is to save bandwidth or storage space.
It makes sense for streaming or portable devices.
 
However it's a proprietary closed-off format that they're using to sell hardware, not an open format being offered for anyone to use.
Until there's an open-source encoder and decoder, the format should be ignored.
Supporting MQA in its current state is anti-consumer.

[dtc]

I don't see the point of MQA.  It seems like a step backward. 

I would want to listen to it before deciding. Some of the early reviews, and yes you have to take reviews with a big grain of salt, are quite positive about how it sounds, even compared to lossless hi rez. And, yes it is a proprietary system which does not auger well for it. But, time will tell.

[mark_h]

I don't see the point of MQA.  It seems like a step backward. 

What do you think MQA is...?

[mark_h]

The point is to save bandwidth or storage space.

That's just a side benefit, it's not the point of MQA...

[Hendrik]

The MQA file format is nothing but a fancy way to compress higher sample rates into otherwise smaller files. Its no secret magic. In concept its similar to the ideas behind HDCD, use an ordinary CD-compatible stream and hide extra data in the low bits of the audio, which should not impact the audio much if played on devices that don't support it. HDCD used this to boost the dynamic range (ie. by extending the bitdepth to 20 bits), and MQA uses that to boost the sample rate instead.

What they market as MQA on the other hand is a lot of different things, from the production side to the playback device, but thats entirely unrelated to the file format and could be achieved with any other file format, if they would want that (which they obviously don't, so they can lock you into their proprietary chain)

Fact remains that without the format being "open", we will not be able to support it either way. Its rather unlikely that we'll start paying MQA to play their format.
If you have a MQA aware DAC, you can play MQA FLAC files today, just need to make sure MC does not modify them in any way.

[mark_h]

The MQA file format is nothing but a fancy way to compress higher sample rates into otherwise smaller files. Its no secret magic. In concept its similar to the ideas behind HDCD, use an ordinary CD-compatible stream and hide extra data in the low bits of the audio, which should not impact the audio much if played on devices that don't support it. HDCD used this to boost the dynamic range (ie. by extending the bitdepth to 20 bits), and MQA uses that to boost the sample rate instead.

Correct.

What they market as MQA on the other hand is a lot of different things, from the production side to the playback device, but thats entirely unrelated to the file format and could be achieved with any other file format, if they would want that (which they obviously don't, so they can lock you into their proprietary chain)

Correct.

Fact remains that without the format being "open", we will not be able to support it either way. Its rather unlikely that we'll start paying MQA to play their format.
If you have a MQA aware DAC, you can play MQA FLAC files today, just need to make sure MC does not modify them in any way.

Well, as you said, MC already supports when set to pass through files unaltered.

I actually don't think there's anything else MC needs to do as an MQA capable DAC is needed at the final point, and MC is never in that position anyway?

But anybody that says that MQA is simply a file packing method is utterly missing what MQA is actually about, which is fundamentally changing (for the better) the way we capture and playback audio.  The file packing is simply a bit of genius on Meridian's part to get their format out there as nobody, not even JRiver, need to do anything other than pass unmolested data to support it.

[Hendrik]

I actually don't think there's anything else MC needs to do as an MQA capable DAC is needed at the final point, and MC is never in that position anyway?

So.. people that keep requesting "MQA support" just don't know any better, and we don't actually need to do anything?
I like it when we don't need to do anything.

[mark_h]

So.. people that keep requesting "MQA support" just don't know any better, and we don't actually need to do anything?

[see post below]

I like it when we don't need to do anything.

MC simply needs to pass the unmolested files onwards to an MQA capable DAC, which it can already do so, yes, MC already supports MQA in that sense.

My main objection is when people dismiss MQA as simply another packing scheme.  That is just one, albeit technically impressive, aspect of MQA.  MQA, as noted above, is actually a fundamental reinvention of audio capture and playback and could well change the way we listen to music.

[mattkhan]

so "passing unmolested" means no DSP?

[mark_h]

Thinking about this some more, I realise that perhaps what people are after is that MC can unpack the MQA, so that even if there isn't an MQA DAC at the end of the chain they still have access to the high res PCM audio embedded within.  In which case, they have a point...

Otherwise the purchase decisions become more difficult - do I buy the vanilla 96/24 PCM or the MQA 44.1 (which has 96/24 embedded), but can only get 44.1 playback in MC.  Or do I have to buy both... or find a software package that supports MQA?

Can we get MQA support in MC please?!  

[mark_h]

so "passing unmolested" means no DSP?

Most likely, but depends on exactly how MQA unpacks, ie if it is expected to use the embedded MQA info to "mix" with the original audio, and you have changed the original audio via DSP then the output is going to be what?

[blgentry]

I'm of the opinion that MQA is a gimmick format.  It's LOSSY compressed.  Who needs lossy compression at this time in computer/digital audio?

I don't doubt that there are mastering standards that go along with this.  But why are you saying it's revolutionary and could "change the way we listen to music"?  It's not like they can re-record the millions of titles that already exist.  In that way, it's sort of like DSD:  Great intentions.  *Teeny tiny* little catalog.  Except that it's lossy!  So why would I be interested?

Brian.

[Awesome Donkey]

I don't doubt that there are mastering standards that go along with this.  But why are you saying it's revolutionary and could "change the way we listen to music"?  It's not like they can re-record the millions of titles that already exist.  In that way, it's sort of like DSD:  Great intentions.  *Teeny tiny* little catalog.  Except that it's lossy!  So why would I be interested?

Completely agree with this.

That, and it being a proprietary format with licensing costs. That's the big sticking point, in my mind.

[mark_h]

It's not like they can re-record the millions of titles that already exist.

And yet, part of MQA is to go back to the studio masters, find out what ADC/mics/digital filters etc were used during recording and then CORRECT the original recording limitations as part of MQA encoding process....  Effectively re-recording the millions of titles that already exist

And then on the playback side is the new method of delivering audio accuracy, way beyond what PCM can do, based on newer understanding of human auditory system.

Anybody asking questions hasn't understood what MQA is doing, IMO.

There is plenty of MQA detail out there now, and lots of extremely positive reported listening experiences.

And more and more studios are coming on board.  MQA is coming...

MQA is the real deal.

[Awesome Donkey]

I still have doubts about MQA. And I won't personally bother with MQA - I have my lossless FLAC library which consists of music from 16/44.1 up to 24/192, which is perfect.

[mattkhan]

Most likely, but depends on exactly how MQA unpacks, ie if it is expected to use the embedded MQA info to "mix" with the original audio, and you have changed the original audio via DSP then the output is going to be what?

I don't follow, possibly because I am not au fait with the details of MQA. Once you've decoded it then you have your final output and then DSP (which could mean active crossovers/bass management or room/speaker correction) is applied on top of that.

And then on the playback side is the new method of delivering audio accuracy, way beyond what PCM can do, based on newer understanding of human auditory system.

any links summarising this?

[mark_h]

I don't follow, possibly because I am not au fait with the details of MQA. Once you've decoded it then you have your final output and then DSP (which could mean active crossovers/bass management or room/speaker correction) is applied on top of that.
any links summarising this?

DSP after decoding might be possible, in a system that is fully MQA compliant architecture, eq Meridian's own(!) as MQA needs new hardware as well, which is going to be the biggest hurdle for most, myself included.  MQA decoded by MC would be turned into PCM, not MQA, but you'd be free to DSP as normal, but you would lose the ultimate benefits of MQA, while gaining from the encoding improvements noted above.

[mark_h]

I have my lossless FLAC library which consists of music from 16/44.1 up to 24/192, which is perfect.

Not according to Meridian

[mark_h]

A great primer:

https://www.youtube.com/watch?v=T5o6XHVK2HA

[Hendrik]

A great primer:

https://www.youtube.com/watch?v=T5o6XHVK2HA

That video really doesn't explain much why the MQA format is required (hint: because it isn't), it mostly focuses on how the mastering process was improved, the ADCs and filters involved.
FLAC at 192/24 or so could deliver the exact same quality, given the same mastering process, but then they wouldn't be able to sell you new hardware as well.

Of course this FLAC would be bigger, but not less quality.

Its important to keep in mind that the MQA format is in fact a lossy compression, you can't hide extra data in a normal FLAC signal without affecting the original data in some way, which I find hilarious.
For years everyone screams for lossless, now someone invents a new lossy compression, and suddenly its the best thing ever.

So in short, great, let them improve the mastering process. But trying to sell new hardware and locked down formats to market this is just money making, not an advance in audio.

MC does allow for third-party input plugins though, so anyone could provide one.

[flac.rules]

A great primer:

https://www.youtube.com/watch?v=T5o6XHVK2HA

That video says very little. The whole "time smearing"-argument seems not to be backed by actual listening tests, and you can fix it wth higher sampling-rate on regular files is desired. There is some talk about analyzing the audio to compensate for faults in studio-equipment, but I am very skeptical to what effect they can do this. The compression part is pretty uninteresting.

[dtc]

That video says very little. The whole "time smearing"-argument seems not to be backed by actual listening tests, and you can fix it wth higher sampling-rate on regular files is desired. There is some talk about analyzing the audio to compensate for faults in studio-equipment, but I am very skeptical to what effect they can do this. The compression part is pretty uninteresting.

In the video he discussed the right hand on a piano and says one mqa recording  is the best he has heard. So, yes, he does discuss listening test as they relate to the time smearing issue .

He also does a second session on listening, where he does more listening tests. He is suitably cautious, but his initial impressions are positive.

https://www.youtube.com/watch?v=T5o6XHVK2HA

[mark_h]

That video really doesn't explain much why the MQA format is required (hint: because it isn't)

Based on what, your opinion?

FLAC at 192/24 or so could deliver the exact same quality

No, I don't believe this is the case.  MQA takes impulse timing beyond what PCM is capable of, and closer to what our ears are capable of.  I'm trying to find a citation for you...

Its important to keep in mind that the MQA format is in fact a lossy compression

Based on what evidence?

[mark_h]

That video says very little. The whole "time smearing"-argument seems not to be backed by actual listening tests,

Backed not only by listening but by science :p

and you can fix it wth higher sampling-rate on regular files is desired.

No, because there is a fundamental timing limitation with PCM that needs to be improved upon, which is what MQA does.  But it does require new hardware to fully achieve.

[dtc]

MQA is certainly a controversial topic. Some think it is complete snake oil and some think it is the next great think for music playback. The result will undoubtedly be somewhere in the middle. From reading the brief comments here, there is more skepticism than support, which is fine. However, I really hope people will read more about the format and follow its roll out. If it has potential for sound improvement, it needs to be considered. Meridian has been a strong player in the digital playback area for a long time and their latest effort should not be simply dismissed before more is know about the potential sound improvements. Stereophile and Absolute sound have both written about it recently and the audiophile sites have lots of discussion about it.

We will never resolve the MQA issue here. I just hope people do not dismiss it based on partial information, and sometimes wrong information, and before we fully understand it potential benefits as the roll out continues.

Separately, both Tidal and Roon are embracing  MQA. As I understand it, Roon is planning a software decoder, not just pass through. 2L has MQA files available for download.

Again, time will tell how this all works out. I, for one, will keep an open mind until more is known.

[flac.rules]

Backed not only by listening but by science :p

No, because there is a fundamental timing limitation with PCM that needs to be improved upon, which is what MQA does.  But it does require new hardware to fully achieve.

What science?

Why does it need improving? On what grounds can one presume that this "need to be improved", what listening test show that this i necessary?

The info coming from Meridian/MQA themselves and Bob Stuart is also disturbing. The claims are sometimes pretty outlandish, which makes their claims in general not very trustworthy.

[flac.rules]

MQA is certainly a controversial topic. Some think it is complete snake oil and some think it is the next great think for music playback. The result will undoubtedly be somewhere in the middle. From reading the brief comments here, there is more skepticism than support, which is fine. However, I really hope people will read more about the format and follow its roll out. If it has potential for sound improvement, it needs to be considered. Meridian has been a strong player in the digital playback area for a long time and their latest effort should not be simply dismissed before more is know about the potential sound improvements. Stereophile and Absolute sound have both written about it recently and the audiophile sites have lots of discussion about it.

We will never resolve the MQA issue here. I just hope people do not dismiss it based on partial information, and sometimes wrong information, and before we fully understand it potential benefits as the roll out continues.

Separately, both Tidal and Roon are embracing  MQA. As I understand it, Roon is planning a software decoder, not just pass through. 2L has MQA files available for download.

Again, time will tell how this all works out. I, for one, will keep an open mind until more is known.

I highly doubt the result will be "somewhere in the middle", why should believe something like that? The lack of info is solely because of Meridian controlling the info. As I see it the format has a bad for customers business model, is unnecessary to obtain the claimed gains, and the gains (if any) will be on the outer fringe of noticability for humans.

[dtc]

I highly doubt the result will be "somewhere in the middle", why should believe something like that? The lack of info is solely because of Meridian controlling the info. As I see it the format has a bad for customers business model, is unnecessary to obtain the claimed gains, and the gains (if any) will be on the outer fringe of noticability for humans.

OK - not necessarily "in the middle" but between the end points. Happier?

You clearly have made up your mind without even listening. I will wait and listen. If it turns out to be a big improvement will you listen to it? By the way, what some perceive as the "outer fringe" in clearly discernible  to others.

[Hendrik]

Based on what evidence?

Its quite obvious if you read the technical details of how it works, and the video you quoted even says so.

[Awesome Donkey]

This blog entry caught my attention: http://benchmarkmedia.com/blogs/application_notes/163302855-is-mqa-doa and here's full technical details about MQA: http://www.aes.org/e-lib/browse.cfm?elib=17501

Anything considered "lossy" will not touch my PC and music library, regardless of the 'benefits' it claims, ever.

[mattkhan]

http://archimago.blogspot.co.uk/2016/02/measurements-impressions-meridian.html is another sceptical piece

I'm firmly in the "the speakers and room make a much bigger difference" camp so if it meant no DSP then no interest from me (until there is a software decoder anyway)

[mark_h]

What science?

You will have to go digging for Bob Stuart's papers on the subject.

Why does it need improving? On what grounds can one presume that this "need to be improved", what listening test show that this i necessary?

You will have to go digging for Bob Stuart's papers on the subject.

The info coming from Meridian/MQA themselves and Bob Stuart is also disturbing. The claims are sometimes pretty outlandish, which makes their claims in general not very trustworthy.

Bob Stuart, and Meridian, are among the most highly regarded individuals and companies in the business, with proven track records.  To suggest their claims are outlandish is, outlandish :p

[ferday]

Backed not only by listening but by science :p

No, because there is a fundamental timing limitation with PCM that needs to be improved upon, which is what MQA does.  But it does require new hardware to fully achieve.

yes, science...

just what we need is a "higher resolution" version of a brickwalled pile of garbage recording.  sharper transients at 0dbfs!  sounds grand.  

[mark_h]

Its quite obvious if you read the technical details of how it works, and the video you quoted even says so.

Ok, so the MQA bits are buried deep down in the least significant bits of the PCM, making the PCM wrapper lossy, but is it *audibly* lossy or is this just scemantics?  

[mark_h]

We may never settle the MQA debate, but what if the market does and MQA is adopted widely?  Will MC support it then?

[mark_h]

I'm firmly in the "the speakers and room make a much bigger difference" camp so if it meant no DSP then no interest from me (until there is a software decoder anyway)

I certainly agree with that.  I cannot imagine running a system without digital room correction, for instance, and if MQA precludes that then I will not adopt it until I have such a solution available.

[JimH]

We may never settle the MQA debate, but what if the market does and MQA is adopted widely?  Will MC support it then?

Popularity doesn't determine quality.

We follow the path that we believe is true, and not always one that many choose.

We don't run barking after every car that passes.

[Awesome Donkey]

And like Hendrik said, a developer could create a third-party input plugin with support.

[flac.rules]

You will have to go digging for Bob Stuart's papers on the subject.

You will have to go digging for Bob Stuart's papers on the subject.

Bob Stuart, and Meridian, are among the most highly regarded individuals and companies in the business, with proven track records.  To suggest their claims are outlandish is, outlandish :p

I have seen the technical rundown on this, it is not very convincing, the papers have no objective listening test as far as I known.

Really? Proven track record or not, many of the claims are clearly outlandish, just look at the stupid "quality-graph" posted here from the website, it is at the very best misleading.

[mwillems]

That quality chart is hilarious.  Reel to Reel tape (with all of the absurd noisefloor and compression problems that come with tape) is the highest quality existing medium?  Vinyl is higher quality than any digital medium even though the source material for most modern music is already digital and vinyl playback necessarily adds distortion?  CD's are higher quality than downloads or streaming, even though you can get downloads that are a lossless copy of the CDs?  Huh?

Not to be unpleasant, but that chart isn't just misleading, it's simply wrong for any objective value of the word "quality".  It looks like a cyncial attempt to encapsulate people's preconceived notions about audio, and suggests that MQA is a marketing ploy rather than an attempt to try to advance the state of the art of audio.  And of course when you tell people what they already believe, you'll be hailed for "speaking the truth".  

Maybe there's something there, but their marketing materials are not particularly confidence inspiring.

[mojave]

That quality convenience chart is hilarious.

And how are LP's more convenient than DVD-A/SACD? DVD-A/SACD actually worked in any CD or DVD player (although all you got were the normal CD/DVD tracks). There were also DVD-A/SACD players for the car. You can buy an OPPO player that plays both formats. With an LP you need a preamp, turntable, cartridge, arm, etc.

[RD James]

No, I don't believe this is the case.  MQA takes impulse timing beyond what PCM is capable of, and closer to what our ears are capable of.  I'm trying to find a citation for you...

They're claiming 10µs of "time smearing" which, based on this graphic touting the advantages of 1x DSD having only 3µs "time smearing", would mean that you have transient information roughly equivalent to an 846.72 kHz sample rate.
 
So I'm not sure how it revolutionizes anything in the time domain when JRiver already supports 8x DSD.
 
And I don't believe what they're claiming about transient information that short being applicable to music. To hear it, your would have to be capable of hearing frequencies more than an order of magnitude higher than the scientifically accepted range for human hearing.
Good luck finding some kind of speaker capable of playing it too.
 

That quality chart is hilarious.  Reel to Reel tape (with all of the absurd noisefloor and compression problems that come with tape) is the highest quality existing medium?  Vinyl is higher quality than any digital medium even though the source material for most modern music is already digital and vinyl playback necessarily adds distortion?  CD's are higher quality than downloads or streaming, even though you can get downloads that are a lossless copy of the CDs?  Huh?

Not to be unpleasant, but that chart isn't just misleading, it's simply wrong for any objective value of the word "quality".  It looks like a cyncial attempt to encapsulate people's preconceived notions about audio, and suggests that MQA is a marketing ploy rather than an attempt to try to advance the state of the art of audio.  And of course when you tell people what they already believe, you'll be hailed for "speaking the truth". 

Maybe there's something there, but their marketing materials are not particularly confidence inspiring.

Agreed. It's pretty clear that they're marketing to audiophiles, not anyone that is looking at this with a critical eye.
 

And how are LP's more convenient than DVD-A/SACD? DVD-A/SACD actually worked in any CD or DVD player (although all you got were the normal CD/DVD tracks). There were also DVD-A/SACD players for the car. You can buy an OPPO player that plays both formats. With an LP you need a preamp, turntable, cartridge, arm, etc.

They're also claiming that downloads are lower quality than DVD-A (24/96 PCM) or SACD (1x DSD)

[mark_h]

That quality chart is hilarious.

Don't let the convenience chart become a convenient straw man

[mark_h]

A good Q&A with Bob, containing lots of details and, at the bottom, citations to relevant papers.

http://www.computeraudiophile.com/content/694-comprehensive-q-mqa-s-bob-stuart/

[flac.rules]

Don't let the convenience chart become a convenient straw man

No, but a company that presents itself with BS-charts does loose credibility.

[flac.rules]

A good Q&A with Bob, containing lots of details and, at the bottom, citations to relevant papers.

http://www.computeraudiophile.com/content/694-comprehensive-q-mqa-s-bob-stuart/

Yeah, I have read it, it is one of the reasons I do not belive this format makes much of a difference (to perceived sound quality)

[dtc]

And how are LP's more convenient than DVD-A/SACD? DVD-A/SACD actually worked in any CD or DVD player (although all you got were the normal CD/DVD tracks).

Every CD player did not read DVD-A and SACD data. You needed a specially designed one to do so. That was one of the problem with the formats. Everyone had to buy a new player. Hybrid SACDs had a redbook layer on them, but that was a later development and only provided a 16/44 data. The point of the chart was that everyone had to replace there CD players to read the new formats, hence the low convenience factor.

[mark_h]

Have any of the naysayers heard MQA yet?