Jump to content

MQA Users & Discussion Thread


Guest AndrewC

Recommended Posts

12 hours ago, Rob181 said:

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

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

As republican Kim Beasley said when accepting the vice-regal position in WA, it is not impossible to separate out these things!

Link to comment
Share on other sites



On 08/04/2018 at 9:30 PM, rmpfyf said:

there is nothing MQA can do that PCM doesn't. This is basic signal processing

Aside from 'proprietry' processing which MQA may do ... that was not be done on the vanila PCM side.

 

Eg. compensating for ADC behaviour in recording, or compensasting for DAC behaviour in playback.

 

... which I don't think is particularly a good thing (we should let converter manufactures improve things, rather than handing over control to a central party who becomes "response for quality") .....   but it's possible.

Link to comment
Share on other sites

  • Volunteer
12 minutes ago, davewantsmoore said:

Yes, of course I did.

 

If someone claims that "everyone is wrong", then it's hard not to ignore.

you are a far more patient man that I am :)

I found him to be ridiculously wordy and repetitive and full of himself

Link to comment
Share on other sites



On ‎4‎/‎9‎/‎2018 at 5:25 PM, Rob181 said:

Kim was referring to his cost benefit analysis...

Eat some humble pie (or in Kims case the whole pie) for 40 pieces of silver...

Or in the modern veracular...a truck load of dough...

As with Bob Stuart, I very much doubt that Kim needs the money - perhaps fame (or a lasting legacy) rather than fortune, with a dash of altruism (doing good for the community) thrown in.

  • Love 1
Link to comment
Share on other sites

1 hour ago, legend said:

As with Bob Stuart, I very much doubt that Kim needs the money - perhaps fame (or a lasting legacy) rather than fortune, with a dash of altruism (doing good for the community) thrown in.

"Always look on the bright side of life..."

Link to comment
Share on other sites

On ‎4‎/‎9‎/‎2018 at 7:10 AM, legend said:

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.

 

Last night I managed to find time to download the 352 kHz versions of 2L’s Mozart Violin Concerto Allegro – unsure whether it was not there the last time I looked or whether I ignored it because the DEQX DAC (my only one at the time) just goes up to 192 kHz.  It took about 6-7 minutes on my 12 Mb/s Internet connection so given the playing time is around 8-9 minutes I guess it might be streamed ‘live’ – though if no one else is using the connection.

 

Also downloaded the 44.1 kHz (Redbook) versions of original & remixed – too much choice and quite confusing after downloading because of the way the files are named.  And I have now listened to the Allegro so many times that I don’t want to hear it again!

 

My overall conclusions were:

1.  in 44.1 kHz, the original mix sounded less strident and so more listenable than the remixed file.

2.  for both mixes, the 352 kHz and the MQA (which is converted to 352 kHz by my Pro-ject S2 DAC) both sound better (less strident, more detailed, less congested) than the 44.1 Redbook.

3.  for both mixes the 352 kHz MQA version sounds more real/involving/PRAT to me than the 352 kHz PCM version.

4.  with 356 kHz MQA, the 2016 mix sounds better than the original mix – more detailed but more relaxed – contrary to 1. above.

 

The differences are not huge but I think I could distinguish them double blind.  However the conclusions may depend on my speaker playback system which is designed high resolution/low distortion (unfortunately I don’t have a good headphone system to compare); and on my personal SQ preferences which are biased against hardness/stridency and towards openness but detailed as well as PRAT; and on the music being played.

Edited by legend
addition
Link to comment
Share on other sites

Guest Eggcup The Daft
5 hours ago, legend said:

 

Last night I managed to find time to download the 352 kHz versions of 2L’s Mozart Violin Concerto Allegro – unsure whether it was not there the last time I looked or whether I ignored it because the DEQX DAC (my only one at the time) just goes up to 192 kHz.  It took about 6-7 minutes on my 12 Mb/s Internet connection so given the playing time is around 8-9 minutes I guess it might be streamed ‘live’ – though if no one else is using the connection.

 

Also downloaded the 44.1 kHz (Redbook) versions of original & remixed – too much choice and quite confusing after downloading because of the way the files are named.  And I have now listened to the Allegro so many times that I don’t want to hear it again!

 

My overall conclusions were:

1.  in 44.1 kHz, the original mix sounded less strident and so more listenable than the remixed file.

2.  for both mixes, the 352 kHz and the MQA (which is converted to 352 kHz by my Pro-ject S2 DAC) both sound better (less strident, more detailed, less congested) than the 44.1 Redbook.

3.  for both mixes the 352 kHz MQA version sounds more real/involving/PRAT to me than the 352 kHz PCM version.

4.  with 356 kHz MQA, the 2016 mix sounds better than the original mix – more detailed but more relaxed – contrary to 1. above.

 

The differences are not huge but I think I could distinguish them double blind.  However the conclusions may depend on my speaker playback system which is designed high resolution/low distortion (unfortunately I don’t have a good headphone system to compare); and on my personal SQ preferences which are biased against hardness/stridency and towards openness but detailed as well as PRAT; and on the music being played.

Thanks.

 

I know the feeling, once you've listened to that Allegro often enough through headphones in one session, the differences start to melt away... playing files through my DOS switching program (which hides which is A and B by renaming them) I could tell the difference between the two mixes easily though that was because with the Dragonfly, at least, the cellos are placed further to the right in the MQA mix.

 

I preferred the original mix through MQA/Tidal (presumably this is at 356) as well with lower resolution files direct - in my set up the lower end is clearer with the original (that shift again in the cellos).

 

I'm not sure about some of the other differences I "heard" while listening sighted, but I would venture that the solo violin attack was harder in the 2016 mix, especially when listening to the non-MQA versions of that mix.

 

 

Link to comment
Share on other sites



  • Volunteer
22 minutes ago, Eggcup The Daft said:

I preferred the original mix through MQA/Tidal (presumably this is at 356) as well with lower resolution files direct - in my set up the lower end is clearer with the original (that shift again in the cellos).

I'm being thick here but I don't know which one you preferred (sorry)

 

Link to comment
Share on other sites

Guest Eggcup The Daft
6 hours ago, Sir Sanders Zingmore said:

I'm being thick here but I don't know which one you preferred (sorry)

 

If you want preference, I preferred the original mix to the MQA version throughout.

 

I'm not prepared to go between the MQA and the 16/44.1 versions here, as the 16 bit is not available to me on Tidal, meaning I'm using different PC software to play each, and that's not fair. I prefer decoded to non-decoded Tidal Master playback for this, through the Tidal app, keeping exclusive mode on, and that across both versions. The solo violin was more realistic and the strings more delineated when decoded.

 

Via my Oppo 105 into my living room system (no MQA decoding here of course) I recently preferred 352/24 to 96/24 to 16/44.1 from the test bench download, sighted, this is possibly unconcious bias as I don't have anything specific that I can hang that preference on though. That's using the original mix which I downloaded a while ago.

 

I've just compared the new mix 352/24 with whatever Tidal supply to their Oppo app in hifi mode, and there was a big difference, the orchestra neither sounded properly in time or like groups of instruments from Tidal. In fact, the worst I've heard it.

 

A few years ago I worked through the original mix downloads (the only one, then) on my old Quadral bookshelves and heard no difference between any of the files. The rest of the system was the same.

 

I often prefer Tidal to my CDs played through the Oppo, and where I'm reasonably certain the mix is the same, I can't usually tell any difference between Tidal HiFi and ripped CDs played through the Oppo, either from a Seagate Personal Cloud share or from an external desktop hard drive plugged in the back.

 

 

Link to comment
Share on other sites

Since my last post I have been trying to find music other than the 2L recordings with which to compare MQA with alternative formats and where the provenance of each type is likely to be the same. Last night I found some of these by David Elias on the Oppo website (note Oppo Audio has just ceased manufacturing – another straw in the wind for those who think audio is a good business in which to make money!):

https://oppodigital.com/hra/dsd-by-davidelias.aspx

 

I first listened to the track “Vision of Her” with its trio of musicians in DSD64 (in which it was recorded) then its conversion to WAV (PCM) 88.2/24 and finally MQA 352.8/24.  I thought the DSD64 version sounded great with a lovely warmth and openness but slightly fulsome bass.  To me the WAV version sounded a bit tighter in the bass but overall a bit thin and ‘lifeless’.  The MQA version seemed to combine the best features of the previous ones – a nice openness but with tighter bass.  In addition the MQA version seemed to have better attack on the plucking of acoustic guitars and David’s voice sounded a bit more real.

 

Similar results were obtained with the track “Morning Light Western Town” which has more musicians playing and so is a bit more complex.  The track starts with a deep rumble that had more texture with MQA than DSD64 and I could hear the hit on cymbals with their subsequent resonance more clearly with MQA so they sounded more real.  That said I would have been very happy with the DSD64 version had the MQA version not been available. 

 

Again this preference for MQA may be a result of my equipment and SQ preferences - and is just a statistical sample of one!  However as part of this personal exploration of MQA I will continue to try to find different types of music and recording techniques to make the comparison where the province of the different versions is reasonably certain.  If anyone knows of any then please share!  I have plenty of examples of music in hires 88/24 or 96/24 which are also available in MQA on Tidal (that I usually prefer) but although in most cases it seems unlikely that they have been remastered/remixed one cannot be absolutely sure.

Elias DSD64.jpg

Elias MQA352.jpg

Elias PCM88.jpg

  • Like 1
Link to comment
Share on other sites

5 hours ago, legend said:

Again this preference for MQA may be a result of my equipment and SQ preferences

I'm glad you acknowledge it. In my opinion, it is obviously your equipment and preferences. I think we get the gist. You always prefer the mqa every single time. 

 

But the problem I have is that when I record an LP and play it back it sounds just like the LP. This demonstrates to me that there are no audible issues with pcm recording when played back on good equipment.

 

But when I heard mqa from the 2L website and compared it to the 44.1khz non mqa file, the regular PCM file sounded more natural and the mqa file lost attack and clarity. I'm clearly not the only person to hear this due to numerous reports. The timing error introduced by the mqa process, ( as measured by archimago and others) is the likely cause of what I hear.

 

There are many ways to achieve a personally "pleasant" sound. Some use valve amps, others vintage speakers, and some use EQ, and some use NOS DACs, PIO capacitors and the list goes on...

 

I'm not enthusiastic about these methods because they introduce distortions which aren't on the original recording. (Anything not part of the original recording is a distortion)

Mqa modifies the originaly recorded data and so technically, measurably, and intrinsically as part of its process adds distortion.

 

You obviously prefer that.

I don't.

 

 

 

 

 

 

 

 

  • Like 1
Link to comment
Share on other sites

34 minutes ago, eltech said:

The timing error introduced by the mqa process, ( as measured by archimago and others) is the likely cause of what I hear.

Can you reference specifically which bit you're talking about?

 

IMVHO it's a fair stretch for them to call it an "error".    It's just what a different filter looks like. <shrug>

Link to comment
Share on other sites



42 minutes ago, eltech said:

I'm not enthusiastic about these methods because they introduce distortions which aren't on the original recording. (Anything not part of the original recording is a distortion)

Mqa modifies the originaly recorded data and so technically, measurably, and intrinsically as part of its process adds distortion.

 

Saying MQA adds distortion is the same as saying auto-tune adds distortion. At the end of the day if that is what the studios wish to release to the public, it becomes a moot point.

 

42 minutes ago, eltech said:

You obviously prefer that.

 I think it is safe to say Legend prefers MQA, that is true. But it is not true that Legend prefers 'distortion' as you have put it. If you follow his other posts on SNA you would realise he is actually a lot more like you in his approach to audio design. 

 

 

Edited by LHC
  • Like 1
Link to comment
Share on other sites

4 minutes ago, LHC said:

Saying MQA adds distortion is the same as saying auto-tune adds distortion. At the end of the day if that is what the studios wish to release to the public, it becomes a moot point

Not exactly, LHC. As I see it, an effect used in the studio is part of the artists intent, so it is valid.

Whereas an effect like mqa added afterwards is not part of the artists intent.

Link to comment
Share on other sites

16 minutes ago, LHC said:

I think it is safe to say Legend prefers MQA, that is true. But it is not true that Legend prefers 'distortion' as you have put it. If you follow his other posts on SNA you would realise he is actually a lot more like you in his approach to audio design

I guess we have to disagree on this. I've explained that I think any modification of the originally recorded data is a distortion. Therefore if someone prefers that, then so be it.

In my view, it is added distortion.

 

Legend has previously said that through his DAC violins sound steely. That to me is a dead giveaway that his DAC isn't a good performer.

 

In response I suggested he get a better DAC. But it seems to have fallen on deaf ears.

 

I think it's unfair to tar CD quality audio broadly with a negative brush when using poorly performing equipment.

To me it sounds exquisite! I have no issues whatsoever. 

Edited by eltech
Link to comment
Share on other sites

1 hour ago, eltech said:

I'm glad you acknowledge it. In my opinion, it is obviously your equipment and preferences. I think we get the gist. You always prefer the mqa every single time. 

 

 

It is true that in the 3 cases (from 2 different recording companies) where I have been able to compare the same recording in different formats I have so far preferred MQA - but am still looking for other cases.

 

But although I often do, I don't always enjoy recordings in MQA on Tidal.  A few days while searching on Tidal for some classical piano music (always a good test for fidelity) I came across the some Chopin Etudes played by Maurizio Pollini from the 1970s - and they sounded so awful (hard, bright, jangly etc) that I quickly stopped playing.  I have an LP of the Chopin Polonaisen & Nocturns played by the same artist that won the "Grande Prix du Disque" in the 1970s and it sounds fine!

  • Like 1
Link to comment
Share on other sites

46 minutes ago, davewantsmoore said:

Can you reference specifically which bit you're talking about?

 

IMVHO it's a fair stretch for them to call it an "error".    It's just what a different filter looks like. <shrug>

 

To answer your question.

Here are direct quote from archimago's blog

http://archimago.blogspot.com.au/2018/02/musingsmeasurements-on-blurring-and-why.html?m=1

 

You can see the plots on his blog.

 

 

 

"As discussed previously, MQA uses a weak anti-imaging reconstruction filter that will allow quite a bit of distortion to pass beyond the Nyquist frequency.

While we've seen graphs like the ones above already such as when Stereophile measured the Mytek, notice that we have not seen published measurements of group delay introduced by the MQA filter. Voilà, detailed measurements of each of the 4 filters available for the Mytek Brooklyn DAC using a combination of my friend's AP gear and FuzzMeasure on the Mac for group delay plots"

 

 

"The group delay plot confirm that with the linear phase filters, there is no delay between bass, mid, and treble - no "temporal smear". In comparison, both the minimum phase and MQA settings do indeed add a delay to the treble frequencies. Specifically, for the MQA filter, by 18kHz, there is about 40µs delay compared to <1kHz (not as bad as the sharp roll-off minimum phase filter with 100+µs). In other words, the MQA filter itself creates temporal smearing instead of "de-blurring"!"

 

 

Link to comment
Share on other sites



45 minutes ago, LHC said:

 

Saying MQA adds distortion is the same as saying auto-tune adds distortion. At the end of the day if that is what the studios wish to release to the public, it becomes a moot point.

 

 I think it is safe to say Legend prefers MQA, that is true. But it is not true that Legend prefers 'distortion' as you have put it. If you follow his other posts on SNA you would realise he is actually a lot more like you in his approach to audio design. 

 

 

 

As you say I have spent much of my speaker design life trying to reduce distortion = difference between output and input.  The first graph shows I have managed to reduce it to below 0.1% over much of the frequency range in the Kantus that I have used for these comparisons (blue line is THD, others are the harmonics). And the second graph  shows the distortion of the Ncore amps I presently use is around 0.001% over its whole frequency range and does not rise above 1 kHz, unlike many amps.  The Pro-ject DAC I have been using also quotes distortion levels around 0.001%.

 

As I have stated many times on these forums I am very much against distortion masking - using one form of 'euphonic' distortion to mask other forms.  So I rarely listen to LPs these days because their distortion playback is around 1% and am similarly trying to sell my valve amp!

 

I don't hear any distortion or distortion masking when listening to MQA.

 

 

Kantu9 THD.png

NC252MP distoertion @1W.png

Link to comment
Share on other sites

39 minutes ago, legend said:

I don't hear any distortion or distortion masking when listening to MQA.

It's interesting that you don't but I and others do.

Ive never liked delta sigma DACs because they subtly and sometimes not so subtly change the tone. They do this because of their noise-shaping algorithms. When you listen to a delta sigma DAC you are hearing the algorithm. That's what makes one sound different from another. A good multibit DAC doesn't have this problem because there is no noise shaping happening, whatever bit you feed it is what comes out of its resistor ladder. The result is a natural sound in contrast to a processed sound which you get from a delta sigma DAC. 

I'd describe it also as a homogenised sound - not natural. I agree delta sigma measures well, but when you hear a good multibit DAC it should immediately be apparent what you've been missing. Oh, and good multibit DACs measure very well too!

 

In all my listening tests with class D amps they also fail to make the grade. A top notch class AB amp always sounds more natural. 

 

Link to comment
Share on other sites

  • Volunteer
1 hour ago, legend said:

I don't hear any distortion or distortion masking when listening to MQA.

Interesting because according to the CA article, whether you hear it or not there is measurable distortion. 

Assuming this distortion is what some people either like or dislike about MQA, how does that square with your own design objectives ?

Link to comment
Share on other sites

Guest Eggcup The Daft
1 hour ago, legend said:

 

The Pro-ject DAC I have been using also quotes distortion levels around 0.001%.

 

You'll find that distortion level is best case, probably high resolution 24 bit playback. It tells us the output stage is pretty good.

You have no guarantee at all that that figure is achieved with MQA playback. Distortion with digital is dependent on a number of things. If it's 0.001% with MQA playback I'd be surprised.

 

Audibility is, of course, another matter.

Link to comment
Share on other sites

1 hour ago, Eggcup The Daft said:

You'll find that distortion level is best case, probably high resolution 24 bit playback. It tells us the output stage is pretty good.

You have no guarantee at all that that figure is achieved with MQA playback. Distortion with digital is dependent on a number of things. If it's 0.001% with MQA playback I'd be surprised.

 

Audibility is, of course, another matter.

 

Fair comment!

Link to comment
Share on other sites



  • Recently Browsing   0 members

    • No registered users viewing this page.




×
×
  • Create New...
To Top