MQA: A Review of controversies, concerns, and cautions

[Julf]

Based on what you've shown here, however, I would hypothesize that I could be trying this the wrong way, by directly comparing the 96k and 44.1k versions, where it is impossible to know exactly what the exact, proper alignment of the waveforms is in order to do the null test properly or fully. By converting a file from 96k to 44.1k and back again, one creates two 96k versions that can then be "perfectly" compared in a null test.

There are a bunch of tools, such as Audio Diffmaker, that match up the files in time.

 

[tmtomh]

There are a bunch of tools, such as Audio Diffmaker, that match up the files in time.

Thanks! I'm on a Mac, but will check out tools for that.

After my last reply, I tried comparing a 24/96k high-res file and a ripped CD file from a commercial release where both were sourced from the identical mastering (Led Zeppelin I, 2014 reissue).

I converted the CD rip to 24/96 and then ran a null test between it and the true high-res file. First thing that surprised me was that the peak levels were different. Even after level-matching, though, I could not get them to null out. Clearly they are slightly different masterings - something was done to one or the other to modify it a little bit before release. I wonder how many other "identical" high-res and CD versions of official releases have such differences as well?

 

[Julf]

Clearly they are slightly different masterings - something was done to one or the other to modify it a little bit before release. I wonder how many other "identical" high-res and CD versions of official releases have such differences as well?

Indeed. There is a huge commercial incentive to make the "premium" versions sound "better".

 

[tmtomh]

Indeed. There is a huge commercial incentive to make the "premium" versions sound "better".

Indeed. Of course, a secret tweak in the mastering, based on what a particular mastering engineer prefers in his or her studio with his or her monitoring speakers, is so far from a reliable measure of what might actually sound better (or worse) that it seems pointless. Nevertheless, it appears that it happens...

 

[Parzival]

Wow, fascinating - thank you!

I've done this myself in two ways:

1. Commercial 24/96 file resampled and dithered by me in Audacity, then null-tested in Audacity; and
2. Commercial 24/96 file and commercial CD rip of known-identical mastering (e.g. 2014/15 Led Zeppelin reissues), null-tested in Audacity.

My results typically have the difference RMS low, around -80dB, but with higher peaks, and with plenty of the difference in the 20Hz-20kHz region.

Based on what you've shown here, however, I would hypothesize that I could be trying this the wrong way, by directly comparing the 96k and 44.1k versions, where it is impossible to know exactly what the exact, proper alignment of the waveforms is in order to do the null test properly or fully. By converting a file from 96k to 44.1k and back again, one creates two 96k versions that can then be "perfectly" compared in a null test.

I'm going to try that now. Thanks!

EDIT: Holy moly - I did it with Audacity and got the same results as you - difference file shows information there, but inaudible because ultrasonic. Very cool. I don't buy much high-res material, but still, this is going to save me some money and storage space moving forward!

I hope Dolby or DTS come out with their own master hires codecs for 2 channel audio. With most of the boutique companies sticking to age old cd, flac and dsd and the only option being pseudo hires MQA who want to price gouge the entire value chain, it’s high time some proper innovation is seen in this space.

 

[Parzival]

In some interview recently mark zuck’s announced that oculus is working on spatial audio format and a new codec. Rumour has it that apple too is on the road to have the next iPod pro with spatial audio DSP and a new codec. Let’s see if these tech promises materialise.

 

[Cahudson42]

Let’s see if these tech promises materialise.

But watch for DRM to be wormed in..

 

[Julf]

Why would we need a new codec? What is wrong with FLAC?

 

[Parzival]

Why would we need a new codec? What is wrong with FLAC?

Flac above 96khz is not exactly streaming friendly, doesn’t have bandwidth to carry object oriented metadata, also compression technologies have progressed all this while

 

[krabapple]

Flac above 96khz is not exactly streaming friendly, doesn’t have bandwidth to carry object oriented metadata, also compression technologies have progressed all this while

why do you need 'above 96kHz'?

 

[Parzival]

But watch for DRM to be wormed in..

since the world is big time into streaming anyway, DRM becomes intrinsic and a moot point to the format. Shouldn’t matter to the new generation of audiophiles at all.

 

[Parzival]

why do you need 'above 96kHz'?

to capture more transients in the music information without the need to have some esoteric upscaling mathematics to fill the gaps

 

[krabapple]

to capture more transients in the music information without the need to have some esoteric upscaling mathematics to fill the gaps

Hah, right. The old 'redbook lacks sufficient time resolution' canard. It never ends.

Human hearing isn't a FFT.

 

[Parzival]

Hah, right. The old 'redbook lacks sufficient time resolution' canard. It never ends.

Human hearing isn't a FFT.

yeah never ending debate
But the reason this debate will never end is because the prevalent hires definition is only relative to the older generation CD quality.
The human ear can discern timing information and audio resolution nearly 250- 300khz. this remains a niche science with not too many academics and universities devoting enough time to the study.

old guards debating whether higher hires () is required is akin to saying why need 8K, 1080p is good enough. we all need more

 

[mansr]

Flac above 96khz is not exactly streaming friendly, doesn’t have bandwidth to carry object oriented metadata, also compression technologies have progressed all this while

MQA doesn't support higher than 96 kHz sample rate at all. As for compression, FLAC performs very well. It is theoretically possible to do a little better, but only with lots of computational effort. Doing much better is impossible.

 

[Parzival]

MQA doesn't support higher than 96 kHz sample rate at all. As for compression, FLAC performs very well. It is theoretically possible to do a little better, but only with lots of computational effort. Doing much better is impossible.

Of course MQA does support more than 96khz. Tidal has acoustic and classical albums coded at 192khz. But MQA has its limitations and is mired with debates and controversies about whether even it has higher bit information which to my mind is an existential question

 

[Parzival]

compression algorithms have naturally come a long way in the intervening decades, but flac is legacy and there are entry - exit barriers to switch formats. Also the computational power is at an enterprise level and should not be a client side concern. (Logic is similar to Netflix)
An influential powerhouse platform company like google, fb, apple, ARM etc or a major chip manufacturer like Qualcomm / Broadcom is needed to introduce sweeping changes with an adoptable mass appeal.
It is beyond the scope of meridian which is small time and even biggies Sony/Samsung who have a competitive take to this.

 

[mansr]

Of course MQA does support more than 96khz. Tidal has acoustic and classical albums coded at 192khz.

That's a lie. I'm sorry you fell for it.

 

[Parzival]

That's a lie. I'm sorry you fell for it.

I don’t bother much about MQA, it doesn’t matter to me bro

 

[Julf]

to capture more transients in the music information without the need to have some esoteric upscaling mathematics to fill the gaps

yeah never ending debate

The human ear can discern timing information and audio resolution nearly 250- 300khz. this remains a niche science with not too many academics and universities devoting enough time to the study.

No, it can't. What do you think the timing resolution of 44.1/16 is?