MQA Users & Discussion Thread

[eltech]

@LHC

MQA has nothing to do with quality of studio production techniques or choices. It is not an industry wide treaty whereby parties agree to give a certain dynamic range. Please do not try to obfuscate what it is.

[legend]

Just had an upclose look at the Pro-ject S2 DACdisplay (which is quite tiny).  The first photo is for the hires nonMQA 2L file of the Mozart Violin concerto from hard disc showing it is receiving at 176 kHz as expected;  while the other from the MQA version via Tidal showing 352 kHz which is higher than I had expected and may go someway to explaining the greater sense of ease? 

 

Please note I am just feeling my way (hopefully forward) in all this - 'searching for certainty' in the great title of a book on scientific method.  If MQA does lead to better SQ for whatever treaon then as an audiophile I hope it succeeds and does not just get killed off by prejudice or otherwise.

Edited April 8, 2018 by legend
clarification

[eltech]

54 minutes ago, legend said:

176 kHz as expected;  while the other from the MQA version via Tidal showing 352 kHz

Not comparing apples to apples then.

[Guest Sime]

@legend ive been offered that DAC for a good price, are you liking it?

[sir sanders zingmore]

1 hour ago, legend said:

Please note I am just feeling my way (hopefully forward) in all this - 'searching for certainty' in the great title of a book on scientific method

As a scientist I'm sure you are fully aware of the inherent biases that humans bring to experiments. 

I'd expect nothing but the best science would satisfy you that these biases were eliminated in any audio testing ...

[LHC]

2 hours ago, eltech said:

@LHC

MQA has nothing to do with quality of studio production techniques or choices. It is not an industry wide treaty whereby parties agree to give a certain dynamic range. Please do not try to obfuscate what it is.

 

That is true. But what the MQA process does is to standardise and control a segment of the production chain through authentication (if the Kool-aid is to be believed). So if the end result music files has sonic quality issue, it is a little easier to identify who was at fault. IMO this can incentivises better choices in production. 

[LHC]

2 hours ago, eltech said:

If not a boffin then who is "causing" the behaviour? 

You tell me.

Good question. I put out the liner notes of a CD and list various roles involved with production, including artists and writers (obviously), producers and executive producers, engineers and assistant engineers, mixers, engineers for mix, recording studios, mixing studios, mastering studio, production companies, publishing companies, administration of publishing etc. So I don't know who is responsible for the sound that we are suppose to hear on a music file. 

[Guest rmpfyf]

5 hours ago, legend said:

Anyway it allowed me to compare again in a very high resolution/low distortion system Mozart’s Violin Concerto 4 from 2L in both MQA and non-MQA format (I had done so some time ago but with a lower resolution system that included a Dragonfly Red DAC).  The MQA version was from Tidal; the non-MQA version was the hires 176-24 from their website.  I assume that they are from the same 2016 remix since it seems unlikely 2L would make a 3rd version just for Tidal.

 

There are some big assumptions here, including your DAC's inherent capacities to time and filter, which as you know will be affected by sample rate. A more realistic comparison would have concerned MQA and PCM mixes from known, identical provenance at identical sample rates.

 

5 hours ago, legend said:

With the MQA version there was also a greater sense of PRAT – pace, rhythm & timing – something in which I was well schooled when working for Linn!

 

Then you'd well know that (provenance asides) getting some parity in the sample rates would have made a fairer comparison among the filter characteristics that underpin these qualities. 

 

Needs a better test. 

 

3 hours ago, Sime said:

@eltech

 

Mqa is about preserving the masters. 

Maybe if an artist is aware of this, they may decide in the studio to make it better from the get go. 

 

So is higher sample rate PCM, DSD, etc. No artist or mastering team gets into a studio with any attempt to deliver anything less than best.

 

22 minutes ago, LHC said:

But what the MQA process does is to standardise and control a segment of the production chain through authentication (if the Kool-aid is to be believed). So if the end result music files has sonic quality issue, it is a little easier to identify who was at fault. IMO this can incentivises better choices in production. 

 

Not what it does. Doesn't affect production as you suggest.

[Guest Eggcup The Daft]

1 hour ago, legend said:

Just had an upclose look at the Pro-ject S2 DACdisplay (which is quite tiny).  The first photo is for the hires nonMQA 2L file of the Mozart Violin concerto from hard disc showing it is receiving at 176 kHz as expected;  while the other from the MQA version via Tidal showing 352 kHz which is higher than I had expected and may go someway to explaining the greater sense of ease? 

 

Please note I am just feeling my way (hopefully forward) in all this - 'searching for certainty' in the great title of a book on scientific method.  If MQA does lead to better SQ for whatever treaon then as an audiophile I hope it succeeds and does not just get killed off by prejudice or otherwise.

The Allegro is available at the full 352.8 from the 2L testbench, no excuses for using the 176.4 unless the DAC has a limitation like my Dragonfly (which will happily fully unfold the Tidal version but won't play straight PCM above 96/24.

 

For my money, the original (I'm listening through headphones now) is easily superior to the 2016 "MQA" remix - more weight and separation of the lower strings, better sense of the acoustic and the solo passages flow somewhat better without a slight glassy edge, as I hear it. I don't think the remix is fixed to favour MQA - but I would recommend listening to the original before drawing your final conclusions...

 

 

 

[sir sanders zingmore]

On 4/2/2018 at 8:30 PM, Sir Sanders Zingmore said:

So I just compared my CD rip of Tom Waits "The Heart of Saturday Night" with the MQA version via Tidal

 

My DAC doesn't have the blue light thingy so I guess that means Roon is doing the first decompression for me and I'm missing out of the megahertz information that the 2nd unfold gives

 

Anyway, the upshot is that the MQA version is a bit louder. That's it

 

On 4/2/2018 at 10:54 PM, a.dent said:

Roon won't do any unfolding ATM. Still in the pipeline. So really you're comparing FLAC with un-unfolded MQA.

 

On 4/3/2018 at 12:34 AM, Sir Sanders Zingmore said:

 

I'll compare Tidal vs CD tomorrow. 

 

So just had a listen to a number of tracks via Tidal

 

First thing I'd say is that I think there is a difference. It's subtle and I'm not sure I'd pick it blind. It certainly is not the "new dawn of hifi" or whatever guff the music press was saying in their adverts reviews.

 

My subjective opinions tend to agree with this article although far less strongly and long winded…. essentially the sound seems 'softer". Not lower volume, but less dynamic. As if the leading edges and transients have been softened. It sounds less natural to me

 

The linked article says:

The sustain portion of each note's string resonance seemed to begin, without the sound of that string ever having been picked. With MQA obliterating and erasing the sound of the guitar pick transient from its time domain reconstruction of the signal waveform, the physical effect of the resonating string's sustain seemed sonically to begin and emerge without ever having had any physical cause (the picking of that string). Thus, MQA was sonically committing the ultimate non-causal filter heresy, making the bell ring without ever having been struck.

 

I get where they are coming from but it's not nearly as obvious to me as it seems to be for them (but I have cloth ears and a non-resolving system so what do I know)

 

[LHC]

57 minutes ago, Sir Sanders Zingmore said:

My subjective opinions tend to agree with this article although far less strongly and long winded….

That webpage and site has some pretty controversial writings . Yeah you are right, its long winded too. 

[Rob181]

2 hours ago, Sir Sanders Zingmore said:

I'd expect nothing but the best science would satisfy you that these biases were eliminated in any audio testing ...

Hmmm...interesting assumption given the man has to balance science...

With cost benefit analysis whilst meeting target "price points" within his business...

[LHC]

1 hour ago, rmpfyf said:

So is higher sample rate PCM, DSD, etc. No artist or mastering team gets into a studio with any attempt to deliver anything less than best.

 

That is not the impression one gets from reading Mark Waldrep's HD-Audio blogs. In the Quincy Jones vs Michael Jackson Estate court case it was revealed that re-masters of the original are actually very poor in comparison.

 

1 hour ago, rmpfyf said:

Not what it does. Doesn't affect production as you suggest.

 

I am not suggesting MQA walks around with a big stick enforcing quality control on production. I am saying standardisation can lead to better transparency of the whole process. 

 

[Wimbo]

1 hour ago, Sir Sanders Zingmore said:

The sustain portion of each note's string resonance seemed to begin, without the sound of that string ever having been picked. With MQA obliterating and erasing the sound of the guitar pick transient from its time domain reconstruction of the signal waveform, the physical effect of the resonating string's sustain seemed sonically to begin and emerge without ever having had any physical cause (the picking of that string). Thus, MQA was sonically committing the ultimate non-causal filter heresy, making the bell ring without ever having been struck.

Why I never got on the Digital Bandwagon

Edited April 8, 2018 by Wimbo

[sir sanders zingmore]

12 minutes ago, Wimbo said:

Why I never got on the Digital Bandwagon

Just you wait, there'll be a little blue light on your phono stage next. The MQA  curve will replace RIAA and only accredited phono stages will be able to play those records 

[davewantsmoore]

On 05/04/2018 at 10:58 PM, Sir Sanders Zingmore said:

Here's a very long and very repetitive article.

If you can bear to wade through it (I couldn't), I think it says that unless you use a very wide antialiasing / reconstruction filter, you are not doing it properly. 

And MQA doesn't. 

 

http://www.iar-80.com/page170.html

 

 

Oh dear.   He has misinterpreted the charts provided by MQA.   They do no show the filter response being short  (as he points out that wouldn't work well) ..... they show the end to end impulse response being short.   (perfect would be a single vertical line).

 

A response more like the "single vertical line", represents better accuracy in sample reconstruction.

 

 

He's got his undies bunched up quick badly about it too.

 

Sure, there are lots of people out there who should know better who misunderstand this stuff ......  but his quip that "none of the engineers" do (or words to that effect) is pretty <I'm lost for an adjective here>.    If he's the only one in the world who understand this stuff, then he should be being more helpful <sarcasm>.

[rantan]

1 hour ago, Wimbo said:

Why I never got on the Digital Bandwagon

Talking about bandwagons.

I have relatively recently re -joined the vinyl bandwagon . It is certainly an enjoyable one, but it's still a bandwagon when you see twenty something chicks buying $50 TTs in JB along with an armful of vinyl which would amount to 20 times the cost of the TT.

 

Edit : apologies for the quick OT.

Edited April 8, 2018 by rantan

[sir sanders zingmore]

6 minutes ago, davewantsmoore said:

 

Oh dear.   He has misinterpreted the charts provided by MQA.   They do no show the filter response being short  (as he points out that wouldn't work well) ..... they show the end to end impulse response being short.   (perfect would be a single vertical line).

 

A response more like the "single vertical line", represents better accuracy in sample reconstruction.

 

 

He's got his undies bunched up quick badly about it too.

 

Sure, there are lots of people out there who should know better who misunderstand this stuff ......  but his quip that "none of the engineers" do (or words to that effect) is pretty <I'm lost for an adjective here>.    If he's the only one in the world who understand this stuff, then he should be being more helpful <sarcasm>.

You didn't actually read the whole thing did you ?

[eltech]

2 hours ago, LHC said:

re-masters of the original are actually very poor in comparison

Yes. It's a remaster. You should read the article on gearslutz that has some of the engineers talking about their time in the studio with MJ, they describe how MJ was personally involved with the mixing, and how it took over 200 mixes to get Billy Jean right. (Spoiler- they used the first mix) ha! 

[eltech]

5 hours ago, LHC said:

Good question. I put out the liner notes of a CD and list various roles involved with production, including artists and writers (obviously), producers and executive producers, engineers and assistant engineers, mixers, engineers for mix, recording studios, mixing studios, mastering studio, production companies, publishing companies, administration of publishing etc. So I don't know who is responsible for the sound that we are suppose to hear on a music file. 

Well that is exactly what you get on most CDs. Try reading the liner notes sometimes.

[Guest rmpfyf]

3 hours ago, LHC said:

That is not the impression one gets from reading Mark Waldrep's HD-Audio blogs. In the Quincy Jones vs Michael Jackson Estate court case it was revealed that re-masters of the original are actually very poor in comparison.

 

That's kinda my point.

 

3 hours ago, LHC said:

I am not suggesting MQA walks around with a big stick enforcing quality control on production. I am saying standardisation can lead to better transparency of the whole process. 

 

It's not a standard process. It's PCM with some encoding and compromises FFS, can we all get over it. If you're MQA 'folding' or whatever word we're using for it this week starts from a e.g. 384/24 master, then you will be able to get quality using MQA in a PCM or FLAC container at lesser sample rate that maybe approximates whatever it was mastered from with some different filtering which may or may not be adaptive. It has to be mastered from something which is the 'standard' and if you're capable of playing that back then MQA anything will not beat it.

 

If the master itself was recorded using MQA tools - which would take having a studio seriously believe that that the insane rates they sample at can't realise time domain behaviours that capture art as intended correctly (i.e. that an entire industry has taken a nap on this one and that Bob Stuart and friends alone have a leading clue here) then sure, you could have a handle on an end-to-end MQA process. Though all it would give you (theoretically) is some shorter time-domain response behaviours which would be remedied by - you guessed it - sampling higher again, which you'd have to do for MQA anyway, which kind of makes this end-to-end 'whole process' argument null and void i.e. there is nothing MQA can do that PCM doesn't. This is basic signal processing.

 

The notion of 'transparency' would involve anyone at any stage of the process being able to peek at what's actually happened to the music having gone through an MQA 'folding' process and interrogating as much but a-ha, you can't do that unless you've paid an MQA license. And if you want to verify what the analogue output looks like you need to pay MQA a fee to unfold, your device manufacturer needs to pay a fee to license that too (har har) and another fee to have your DAC ramped and stamped by MQA themselves if you want 'better transparency of the whole process' via little blue light. 

 

So many people need to pay up here that the notion of 'transparency of the whole process' that the odds of the 'whole process' being MQA are small. And what does it mean? How come a $500 or $5000 DAC both get a little blue light? Are they the same? Am I getting the same experience? 

 

Here's MQA licensing in a single equation - it's somewhat heavy around streaming:

 

Bandwidth costs in distribution previously - bandwidth costs in distribution with MQA < MQA licensing fees*

*Customers and content creators wear the rest - sucks if you're a content creator that doesn't distribute, great if you're an artist that owns Tidal

 

Transparent enough a process?

Edited April 8, 2018 by rmpfyf

[LHC]

1 hour ago, rmpfyf said:

If the master itself was recorded using MQA tools - ...  then sure, you could have a handle on an end-to-end MQA process.

Lets take that as the starting point for discussion.

 

1 hour ago, rmpfyf said:

The notion of 'transparency' would involve anyone at any stage of the process being able to peek at what's actually happened to the music  ... you can't do that unless you've paid an MQA license.

Lets say all fees are paid up and you get the little blue light at the end of the process.

 

If the final music quality still sucks (in terms of dynamic range etc), then there is only one possible culprit - the people responsible for the recording using the MQA tools. That is what I meant by transparency, there is nowhere to hide (if all fees are paid). Bad recordings would be shown for what they are. It is hoped this realisation could incentivises better recording to be made in the future.

[Guest rmpfyf]

@LHC - I get what you're suggesting through this is no different to Redbook, just the licensing model is (considerably) more accessible.

 

I'd bet the number of customers that actually care about filtering provenance and wouldn't ever play with their DACs filter options on hires are small. Let alone that there are other 'culprits' as you call them, or everyone with an MQA-compatible anything can turf it in a trash can for the cheapest MQA-enabled thing going because it should sound the same, right?

 

No? There are analogue differences post? Output stages, amps, speakers, acoustics etc? How do they compare to putting a lid on the in-to-out-of-digital bits? Or to tweaking the filter/upsample/higher freq/etc/whatever?

 

Do you only get a blue light when it's end-to-end? A bit from the end? Louder? Tweaked?

 

Get all you're saying here, simply that despite what's suggested what's possible - which I've suggested before is technically significant - is a low probability. Not least because MQA isn't going to make a ton of money off audiophiles initially, there's more from streaming.

 

And really, the notion of MQA tools in recording is fanciful. What technical merit would they add that couldn't be achieved by existing means?

[legend]

15 hours ago, Sime said:

@legend ive been offered that DAC for a good price, are you liking it?

Apologies for the delay in replying but I was called away on domestistica.

 

Yes I very much like the Pro-ject S2 DAC both for its SQ and facilities.  It  is very open with plenty of detail  resolution - two of my top priorities in SQ.  And it has SPDIF & toslink inputs so I can connect it my Sony SACD & Marantz universal player - something that many USB DACs don't do these days.  It has the ability to select various digital filters though I haven't had time to play with these yet.  For around $500 it seems very good value.  But the main reason I bought it was to further investigate MQA - while also good for the money (around $125) I thought my Dragonfly Red DAC that also does MQA was a bit compromised in SQ.

 

There are plenty of reviews to be found by googling - two of the most thorough (and very recent) are

https://www.audioadvice.com/content/2018/02/22/pro-ject-pre-box-s2-digital-review/

https://darko.audio/2018/03/pro-jects-densely-packed-s2-digital/

[legend]

14 hours ago, Eggcup The Daft said:

The Allegro is available at the full 352.8 from the 2L testbench, no excuses for using the 176.4 unless the DAC has a limitation like my Dragonfly (which will happily fully unfold the Tidal version but won't play straight PCM above 96/24.

 

For my money, the original (I'm listening through headphones now) is easily superior to the 2016 "MQA" remix - more weight and separation of the lower strings, better sense of the acoustic and the solo passages flow somewhat better without a slight glassy edge, as I hear it. I don't think the remix is fixed to favour MQA - but I would recommend listening to the original before drawing your final conclusions...

 

 

 

Thanks for the info.  I have not visited the 2L website for a while so will try to download the Allegro & do the comparison later today if time permits.