• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required as is 20 years of participation in forums (not all true). Come here to have fun, be ready to be teased and not take online life too seriously. We now measure and review equipment for free! Click here for details.

MQA Sounds Really Good!

DonH56

Technical Expert
Technical Expert
Forum Donor
Joined
Mar 15, 2016
Messages
3,814
Likes
5,481
Location
Monument, CO
A little bit off topic.

Which is the higher resolution audio format?
96khz/16 bit or 48khz/24 bit.
Depends on how you define "resolution". 96 kHz offers twice the signal bandwidth, up to 48 kHz compared to 24 kHz, but 24 bits offers 256 times the amplitude resolution (24-16 = 8 bits, 2^8 = 256). In terms of SNR 16 bits is about 98 dB max and 24 bits is about 146 dB so 24 bits offers much greater (theoretical) dynamic range. Since 48 kS/s already offers more bandwidth than I can hear I'd choose 24 bits over the higher sampling rate.

Whether both exceed the bandwidth and dynamic range we can use in the real world is always open for debate.
 

KozmoNaut

Active Member
Joined
Sep 28, 2019
Messages
288
Likes
486
Depends on how you define "resolution". 96 kHz offers twice the signal bandwidth, up to 48 kHz compared to 24 kHz, but 24 bits offers 256 times the amplitude resolution (24-16 = 8 bits, 2^8 = 256). In terms of SNR 16 bits is about 98 dB max and 24 bits is about 146 dB so 24 bits offers much greater (theoretical) dynamic range. Since 48 kS/s already offers more bandwidth than I can hear I'd choose 24 bits over the higher sampling rate.

Whether both exceed the bandwidth and dynamic range we can use in the real world is always open for debate.
I know I certainly wouldn't want to experience the playback of a recording making full use of 24 bits of dynamic range, played at a level that makes that entire dynamic range necessary.

I mean, I would certainly be impressed by the aftermath, but I wouldn't like to be there for the actual playback event.

(Figure 30dB background noise, 144dB of dynamic range means peaks of 173dB, or the equivalent of a stun grenade or a .30-06 being fired 1m from your unprotected ear)
 

DonH56

Technical Expert
Technical Expert
Forum Donor
Joined
Mar 15, 2016
Messages
3,814
Likes
5,481
Location
Monument, CO
I know I certainly wouldn't want to experience the playback of a recording making full use of 24 bits of dynamic range, played at a level that makes that entire dynamic range necessary.

I mean, I would certainly be impressed by the aftermath, but I wouldn't like to be there for the actual playback event.

(Figure 30dB background noise, 144dB of dynamic range means peaks of 173dB, or the equivalent of a stun grenade or a .30-06 being fired 1m from your unprotected ear)
I doubt many real-world devices exceed 120 dB or so but the extra dynamic range is nice. Pros appreciate the extra dynamic range when recording, mixing, and mastering as it gives them more headroom to mix with less fear of clipping or losing low-level details.

Hunters in the field have to deal with the rifle blast far closer than 1 m (though not at the muzzle end, hopefully). Probably why one ear is a little worse than the other for a lot of us (hunters). At least we can wear ear protection on the range.

In any event my comments about usage and preference are superfluous to answering the question of resolution so I suppose I should have left them off.
 

KozmoNaut

Active Member
Joined
Sep 28, 2019
Messages
288
Likes
486
The extra dynamic range makes life a lot easier when recording, mixing and mastering, but it really doesn't have any benefits for music that is properly mastered for CD quality, ie. not mastered at -60dB average or something pathologically silly like that.
 

Soniclife

Major Contributor
Forum Donor
Joined
Apr 13, 2017
Messages
2,400
Likes
2,054
Location
UK
Steve jobs eliminated DRM/copy protection from audio industry. It was done 10+ years ago. I don't know how people can be so out of touch with the music industry to accept or make such claims.
These arguments were valid back in 1990s. Not now.
Hmm, it seems that argument is behind the times, the vast majority of revenue now comes from DRM wrapped music.
The reason is that people stand up and shout, they make their displeasure heard and they spread the word far and wide about the scammy nature of MQA, as well as other proprietary formats and DRM schemes. If we just laid down and let the industry bigwigs do whatever they liked, music would be locked down and only temporarily licensed to you. They only see music as a revenue stream, not as culture and art. We music lovers have to stand up to them every time they come up with a new scheme to lock down music.
Then start protesting streaming, they have already stopped issuing some new music on CD, it's only going to increase.
 

Soniclife

Major Contributor
Forum Donor
Joined
Apr 13, 2017
Messages
2,400
Likes
2,054
Location
UK
Watermarking seems to be real... Can one watermark MQA ?
Technically I'm sure the answer is yes. It can survive lossy so no reason it won't survive MQA that I can think of.

I've not found any MQA to have watermarking on it, and I've looked.

One of their main claims would be destroyed if it was watermarked, and open to legal challenge, how could they defend it as master quality if it had watermarking applied after the mastering quality. My cynicism say's it only a matter of time before it happens though.
 

solderdude

Major Contributor
Joined
Jul 21, 2018
Messages
4,122
Likes
7,560
Location
The Neverlands
The reason for thinking watermarking is most likely not possible is because when watermarking the actual audio content is changed.
This means manipulation of all bits can happen.
When one would do this with MQA file the encoded lower bits would change (and thus become corrupt as it is scrambled) or they would have to be smart enough to only modify the first manipulative bit (depends on bit depth).
 

Soniclife

Major Contributor
Forum Donor
Joined
Apr 13, 2017
Messages
2,400
Likes
2,054
Location
UK
The reason for thinking watermarking is most likely not possible is because when watermarking the actual audio content is changed.
This means manipulation of all bits can happen.
When one would do this with MQA file the encoded lower bits would change (and thus become corrupt as it is scrambled) or they would have to be smart enough to only modify the first manipulative bit (depends on bit depth).
It's about 50db down in my tests, so should survive. It won't have to survive thought the whole track either, as I assume it's repeated data. If there was a problem with it being at 50db down they could change it to 40db, etc. It sounds like a dial up modem when isolated.
 

KozmoNaut

Active Member
Joined
Sep 28, 2019
Messages
288
Likes
486
Then start protesting streaming, they have already stopped issuing some new music on CD, it's only going to increase.
I absolutely do, after having used Spotify for a while, and after realizing how little care is put into the quality of their metadata and how little care is given about albums suddenly being unavailable, and then maybe popping back up again as a completely new entry, which means it's still missing from your playlists and collections. Or even worse: replaced with a crappy remastering hackjob.

I held out for a while, because streaming is certainly convenient for casual listening to a huge variety of music. It just isn't very good for building a relatively well-managed collection.

And of course the DRM and user-unfriendly UI changes on top of that.
 

RayDunzl

Major Contributor
Central Scrutinizer
Joined
Mar 9, 2016
Messages
8,188
Likes
5,335
Location
Riverview FL
SoundCloud identified, and blocked, assumedly by automation, my track from Sgt Pepper that I had remastered as an example using 1 bit of resolution, with any original non-zero sample amplified to full scale,

It let me deposit a short version.

1573496981119.png



Nothing in the DRM Detection area will surprise me now.
 
Last edited:

mansr

Addicted to Fun and Learning
Joined
Oct 5, 2018
Messages
895
Likes
2,181
Location
Hampshire
The reason for thinking watermarking is most likely not possible is because when watermarking the actual audio content is changed.
This means manipulation of all bits can happen.
When one would do this with MQA file the encoded lower bits would change (and thus become corrupt as it is scrambled) or they would have to be smart enough to only modify the first manipulative bit (depends on bit depth).
Watermarking would presumably be done before MQA encoding. If applied afterwards, the "authentication" would fail, although the decoder would still output something.
 

LuckyLuke575

Senior Member
Forum Donor
Joined
May 19, 2019
Messages
305
Likes
254
Location
Germany
This is the reasoning I don't get. Keeping it simple, if 24/96, is THE original MQA CAN'T make it better, it is at most a clever/better lossy compressed format. When streaming and storing uncompressed 24/96 is painless MQA has no purpose at all.
It's not a reasoning, it's a statement of my real life experience lol The point is that MQA was used to bring 96/24 to TIDAL, a platform I discovered and liked after buying my first DAC. I have tried Qobuz which has straight PCM which can also be downloaded for offline listening, up to 192/24 which I like, but I found the desktop app to be very buggy, slow, and confusing to use, so I didn't use it after the trial period.

In my mind, the MQA is better quality because of the authentication aspect, where I know there has been a process to check the quality and ensure that the version of the album being played has been verified as being true to the original master. The unfold concept is a bit of a downside as I would prefer to have a direct data stream to the DAC, but I'm very happy with the audio quality on the 'MASTER' albums, so I don't have any practical issues with it.

I can tell you that nothing I've read in this thread would make me change my mind on using MQA for now. I think the value proposition is too good, and I welcome anything that takes music and audio past the Mp3 compressed / lossy hell of the 2000's. So between the High Res concept, and MQA I think they are steps in the right direction to expand high quality audio into the mass market and make it more available which I like.
 

RayDunzl

Major Contributor
Central Scrutinizer
Joined
Mar 9, 2016
Messages
8,188
Likes
5,335
Location
Riverview FL

solderdude

Major Contributor
Joined
Jul 21, 2018
Messages
4,122
Likes
7,560
Location
The Neverlands
Watermarking would presumably be done before MQA encoding. If applied afterwards, the "authentication" would fail, although the decoder would still output something.
Yes, that's what I figured.
 

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
24,353
Likes
46,884
Location
Seattle Area
The reason for thinking watermarking is most likely not possible is because when watermarking the actual audio content is changed.
This means manipulation of all bits can happen.
When one would do this with MQA file the encoded lower bits would change (and thus become corrupt as it is scrambled) or they would have to be smart enough to only modify the first manipulative bit (depends on bit depth).
The only reason to modify the bits post MQA encoding is to use watermarking to embed the ID of the buyer or player. This was attempted years ago in video format whose name escapes me now. And there was a very early music distribution service that did the same. The industry is dead set against doing this and has been for years due to terrible PR it generates.
 

KozmoNaut

Active Member
Joined
Sep 28, 2019
Messages
288
Likes
486
All promo/press kit releases from labels using the bigger promo services (Haulix, iPool, PromoJukeBox and so on) are uniquely watermarked to combat pre-release leaks. None of the reviewers I know have ever complained about a loss of sound quality.

Personally, I haven't been able to detect these particular watermarking methods, but I have absolutely noticed the type used by Universal Music, which plagued Spotify for a long time.
 

Soniclife

Major Contributor
Forum Donor
Joined
Apr 13, 2017
Messages
2,400
Likes
2,054
Location
UK
The industry is dead set against doing this and has been for years due to terrible PR it generates.
Audible watermarking is still heavily used by some of the major labels on streaming content, or are you taking about something else?
 

KozmoNaut

Active Member
Joined
Sep 28, 2019
Messages
288
Likes
486
In my mind, the MQA is better quality because of the authentication aspect, where I know there has been a process to check the quality and ensure that the version of the album being played has been verified as being true to the original master. The unfold concept is a bit of a downside as I would prefer to have a direct data stream to the DAC, but I'm very happy with the audio quality on the 'MASTER' albums, so I don't have any practical issues with it.
Verified as being true to the master the label wants to give you, which isn't necessarily the same as the original studio recordings. Remastering is a thing, you know.

I can tell you that nothing I've read in this thread would make me change my mind on using MQA for now. I think the value proposition is too good, and I welcome anything that takes music and audio past the Mp3 compressed / lossy hell of the 2000's. So between the High Res concept, and MQA I think they are steps in the right direction to expand high quality audio into the mass market and make it more available which I like.
Why not just get lossless CD quality, with no fancy-pants hand waving trickery?
 

digicidal

Major Contributor
Joined
Jul 6, 2019
Messages
1,032
Likes
1,373
Location
Sin City, NV
SoundCloud identified, and blocked, assumedly by automation, my track from Sgt Pepper that I had remastered as an example using 1 bit of resolution, with any original non-zero sample amplified to full scale,
...
Nothing in the DRM Detection area will surprise me now.
That is the real reason that DRM (in the formal definition) is dead (in media)... and it has nothing to do with Steve Jobs at all. With the advent of cloud computing it is a simple matter to "spin up" processor cycles on demand and thus implement analysis algorithms which are applied in-line (or at least near-line) with content distribution (or implement SaaS solutions to do so). Well, that and the public outcry against formal DRM schemes has made many hesitant to employ them liberally.

In large pieces of software/games, which often live in isolated systems of varied capacity, there are still third-party DRM services to provide validation with a minimum of client resources required. With streaming video and audio (and obviously text content) it's a fairly simple matter to analyze in real-time and control distribution based on copyrights (or political correctness, fake news, etc.).
 

Similar threads

Top Bottom