A question about MQA

[rbbert]

If the format is sending the information at 1/2 of 1/3 the file size how is it the same size? I'm reading your post, but not interpeting.

The MQA and a corresponding FLAC (level 8) file of the same original PCM file are about the same size (MQA slightly smaller, because it not completely lossless as FLAC is).

As far as Lumin or Roon or any other software/firmware stating "lossless", this probably only means it is a PCM stream, which by definition is "lossless" whatever the bit depth and sampling rate. To test this, upsample a 16/44.1 file to 24/96, change the amplitude of the file in an audio editor by any small amount (e.g., -0.1 dB, which then makes it "true" 24/96 to an editing program), and see what the Lumin or Roon says.

 

[rbbert]

So why are you using a Wikipedia quote?

This really isn't worth discussing. It has been tested by numerous labs, security agencies and dilettantes; if data-checking is used as part of the copying process, the data will remain unchanged. This is different from rewritable media losing or corrupting data pieces over time; that's why checksums and frequent back-ups are part of any reasonably good data storage.

 

[Kuoppis]

This really isn't worth discussing. It has been tested by numerous labs and dilettantes; if data-checking is used as part of the copying process, the data will remain unchanged. This is different from rewritable media losing or corrupting data pieces over time; that's why checksums and frequent back-ups are part of any reasonably good data storage.

So you as an ER physician are saying that people at the CERN Hadron Collider do not know what they are doing with their data systems? Probably that the CERN report is fake news, maybe?

This is the difference when you are a physician and read about theoretical stuff you most likely do not even understand, nor have a chance of verifying or falsifying. Or if you actually know what you are talking about.

Everyone really in the business knows that data corrupts in many forms and through various factors. It’s unfortunate, but true.

Here’s another one for the Trumpists to obsess about: Did you know that data on your CDs corrupts over time, even though you are not even using them?

Looking forward to the answers [emoji3].

 

[rbbert]

How do the CERN "people" backup their data? I would think the same way financial institutions, government security agencies and other scientific bodies do; copy, data check, checksums.

Here's a simple example for you, relevant to digital audio; take a FLAC file, decode to WAV and save. Or, take a saved WAV file and encode to FLAC. Then, either by running a batch file or doing it manually until you get tired or bored (100x, 1000x, or whatever), decode it to wav and re-encode to FLAC. Decode to WAV a final time, and compare the 2 WAV files using Audio DiffMaker or any other data checking software you choose. Let us know your results.

Data on manufactured CD's rarely corrupts; again, many people have compared the data from 2 old CD's and found no difference. It certainly can happen, that has been documented, just rarely. CD-R's are a different story, in general the physical aspects are not as robust as manufactured CD's.

As you have said, the potential problems with serial copying and data loss and corruption are known and have been known for a long time, and there are a number of methods in common use to correct for that.

 

[Kuoppis]

How do the CERN "people" backup their data? I would think the same way financial institutions, government security agencies and other scientific bodies do; copy, data check, checksums.

Here's a simple example for you, relevant to digital audio; take a FLAC file, decode to WAV and save. Or, take a saved WAV file and encode to FLAC. Then, either by running a batch file or doing it manually until you get tired or bored (100x, 1000x, or whatever), decode it to wav and re-encode to FLAC. Decode to WAV a final time, and compare the 2 WAV files using Audio DiffMaker or any other data checking software you choose. Let us know your results.

Data on manufactured CD's rarely corrupts; again, many people have compared the data from 2 old CD's and found no difference. It certainly can happen, that has been documented, just rarely. CD-R's are a different story, in general the physical aspects are not as robust as manufactured CD's.

As you have said, the potential problems with serial copying and data loss and corruption are known and have been known for a long time, and there are a number of methods in common use to correct for that.

My dear physician, the “CERN people” run one of the most complex IT operations on the planet.

Don’t you guys ever read up on things, before actually answering? Wouldn’t it be nice once in a lifetime to actually know what you are talking about?


Sent from my iPad using Tapatalk

 

[rbbert]

I might add that it is apparent that most companies in the recorded music industry apparently do not take appropriate care of their music storage, digital or analog. Judging from some of what has been permanently lost, backup and storage is irregular and haphazard at best. Private music collectors tend to do a much better job, with secure copying, regular backups and scattered physical storage sites.

 

[steve59]

This thread has gone way past consumer consumer level discussion. Like most format wars, vhs vs beta, hd vs blueray, quality or potential quality have nothing to do with what becomes the standard. We just have to find ways to like what the bean counters decide to feed us.

 

[rbbert]

My dear physician, the “CERN people” run one of the most complex IT operations on the planet.

Don’t you guys ever read up on things, before actually answering? Wouldn’t it be nice once in a lifetime to actually know what you are talking about?

So what is your point (not at all clear from the links you have posted)? You actually think that CERN is more concerned with data storage than the NSA, or Bank of America? CERN is merely idealistic; in this world, unfortunately, money is what makes it go around (a key difference being those groups are not going to let the "outside" world know about it).

Relevant to this audio forum, try the FLAC/WAV experiment I outlined, and you may get a better understanding of how to preserve digital audio data.

 

[rbbert]

...We just have to find ways to like what the bean counters decide to feed us.

This is really the problem with MQA in a nutshell; the powers that be behind MQA would prefer a world where MQA is the only digital audio format with higher digital resolution ("quality") than CD made available to the public.

 

[Kuoppis]

Well, you go ahead an do statistical analysis based on a sample size of one or two. That’s what people always do when they are not able to deduct a reasoning: use anecdotal evidence.


Sent from my iPad using Tapatalk

 

[Kuoppis]

This is really the problem with MQA in a nutshell; the powers that be behind MQA would prefer a world where MQA is the only digital audio format with higher digital resolution ("quality") than CD made available to the public.

Again nonsense: how many times has it been stated here, that me as well as many others, are format agnostic. There are people who just want to enjoy the variety. Always these conspiracy theories where the brain is put out of service.

Vinyl, PCM, DSD, tape - anything goes.


Sent from my iPad using Tapatalk

 

[kzhtoo]

I’m not asking this to be a smart ass and I genuinely do not know the answer to my question. What is MQA so bad for the naysayers apart from MQA guys making a few bucks?

I do have a DAC that can fully unfold MQA files, although MQA was never part of my purchase decision. And I get my MQA from Tidal, and again, without or without MQA has nothing to do with my Tidal subscription. I never felt MQA files sound terribly bad over CD or Hi-Res files and they aren’t also exceptionally sound better than other formats. It really still depends on the particular recording or album. But if MQA helps reduce the bandwidth both for consumers and the streaming companies, why is it so bad about it?

Nothing will eliminate other formats unless the new technology is vastly superior in every way. But meanwhile a new format here and there to address a few gaps here and there, I’m totally okay with.

Note: I also have Quboz subscription.

 

[NekoAudio]

I’m not asking this to be a smart ass and I genuinely do not know the answer to my question. What is MQA so bad for the naysayers apart from MQA guys making a few bucks?

There are several issues with MQA independent of a claim of saving bandwidth. If it had simply been presented as another lossy codec for space savings, like MP3 or Ogg, then we'd be having a very different discussion. But in that case it also would have likely been forgotten by now.

1. The way the technology has been presented and discussed, it leads to misinformation, disinformation, and misunderstandings. As illustrated by this thread.

2. From a purely technical perspective, it is a worse choice than comparable codecs for encoding, storing, and reproducing audio data (not the sound, the data). Similar to vinyl or cassettes being technically inferior to Red Book CDs. You can certainly prefer the sound of X to Y, but that's separate from any technical merits.

3. From a marketing perspective, and as a result of #1, MQA has become a feature requested by consumers that increases the cost and complexity of products. That cost is not free and will be passed on to the consumer when buying hardware, software, and music.

4. The problems MQA advertises itself as solving are problems that it is essentially creating for itself. For example, you ask if MQA helps reduce the bandwidth then isn't that a good thing (independent of any other claims or issues). But the truth is that it doesn't reduce the bandwidth required when compared against alternative codecs, for audio of equal fidelity (e.g. 17-bit/96kHz MQA displayed as 24-bit/96kHz during playback but equal to 17-bit/96kHz LPCM).

5. Unlike multichannel audio, 2-channel consumer audio was the one area unburdened by a proprietary codec. That means anyone could design and build their own product that would be both backwards- and forwards-compatible, and anyone could perfectly transcode audio data constrained only by the technical limits of a codec or medium. That also extends to audio creation. MQA changes that the same way HDMI, HDCP, MPEG-2, Dolby Digital, etc. changed things for multichannel. For video, at least, the industry has made a conscious decision to fight that in the form of the Alliance for Open Media and AV1 codec.

 

[Al M.]

There are several issues with MQA independent of a claim of saving bandwidth. If it had simply been presented as another lossy codec for space savings, like MP3 or Ogg, then we'd be having a very different discussion. But in that case it also would have likely been forgotten by now.

1. The way the technology has been presented and discussed, it leads to misinformation, disinformation, and misunderstandings. As illustrated by this thread.

2. From a purely technical perspective, it is a worse choice than comparable codecs for encoding, storing, and reproducing audio data (not the sound, the data). Similar to vinyl or cassettes being technically inferior to Red Book CDs. You can certainly prefer the sound of X to Y, but that's separate from any technical merits.

3. From a marketing perspective, and as a result of #1, MQA has become a feature requested by consumers that increases the cost and complexity of products. That cost is not free and will be passed on to the consumer when buying hardware, software, and music.

4. The problems MQA advertises itself as solving are problems that it is essentially creating for itself. For example, you ask if MQA helps reduce the bandwidth then isn't that a good thing (independent of any other claims or issues). But the truth is that it doesn't reduce the bandwidth required when compared against alternative codecs, for audio of equal fidelity (e.g. 17-bit/96kHz MQA displayed as 24-bit/96kHz during playback but equal to 17-bit/96kHz LPCM).

All good points. I would add that, as far as I read on the thread at Audiophile Style (former "Computer Audiophile"), on Tidal some titles are only available as MQA anymore. That means the customer is robbed of choice, including even access to Red Book.

5. Unlike multichannel audio, 2-channel consumer audio was the one area unburdened by a proprietary codec. That means anyone could design and build their own product that would be both backwards- and forwards-compatible, and anyone could perfectly transcode audio data constrained only by the technical limits of a codec or medium. That also extends to audio creation. MQA changes that the same way HDMI, HDCP, MPEG-2, Dolby Digital, etc. changed things for multichannel. For video, at least, the industry has made a conscious decision to fight that in the form of the Alliance for Open Media and AV1 codec.

Such an external standard is a main reason why Schiit Audio, to their great credit, has refused to embrace MQA:

Why We Won't Be Supporting MQA

It is also a main reason why the late Charles Hansen (Ayre) has written this essay:

If MQA is the "new world" of audio, I want no part of it - I'd rather dig ditches

 

[Kuoppis]

There are several issues with MQA independent of a claim of saving bandwidth. If it had simply been presented as another lossy codec for space savings, like MP3 or Ogg, then we'd be having a very different discussion. But in that case it also would have likely been forgotten by now.

1. The way the technology has been presented and discussed, it leads to misinformation, disinformation, and misunderstandings. As illustrated by this thread.

2. From a purely technical perspective, it is a worse choice than comparable codecs for encoding, storing, and reproducing audio data (not the sound, the data). Similar to vinyl or cassettes being technically inferior to Red Book CDs. You can certainly prefer the sound of X to Y, but that's separate from any technical merits.

3. From a marketing perspective, and as a result of #1, MQA has become a feature requested by consumers that increases the cost and complexity of products. That cost is not free and will be passed on to the consumer when buying hardware, software, and music.

4. The problems MQA advertises itself as solving are problems that it is essentially creating for itself. For example, you ask if MQA helps reduce the bandwidth then isn't that a good thing (independent of any other claims or issues). But the truth is that it doesn't reduce the bandwidth required when compared against alternative codecs, for audio of equal fidelity (e.g. 17-bit/96kHz MQA displayed as 24-bit/96kHz during playback but equal to 17-bit/96kHz LPCM).

5. Unlike multichannel audio, 2-channel consumer audio was the one area unburdened by a proprietary codec. That means anyone could design and build their own product that would be both backwards- and forwards-compatible, and anyone could perfectly transcode audio data constrained only by the technical limits of a codec or medium. That also extends to audio creation. MQA changes that the same way HDMI, HDCP, MPEG-2, Dolby Digital, etc. changed things for multichannel. For video, at least, the industry has made a conscious decision to fight that in the form of the Alliance for Open Media and AV1 codec.

NekoAudio, a couple of inaccuracies and one thing I do not understand why you state it, as you are an audio dealer.

1. Agreed, MQA marketing has been unclear and a mess.

2. There‘s no reason why MQA should be a worse storage format than any other digital format, it’s still a variant of PCM. As to the other mediums, vinyl is actually technically the superior storage format, as it deteriorates the least. I would guess tape is subject to wear and tear, and maybe some degradation. CDs can start to degrade and lose data already after 10-20 years, audio files on a HDD are susceptible to silent corruption.

3. This is what I don’t understand you saying as a dealer: if MQA is an added feature and increases product price, that means you also make more margin. Why is that bad?

4. You need to differentiate between transmission bandwidth, and playback bandwidth. MQA actually does decrease transmission bandwidth, as the unpacking (de-compression) happens in the DAC.

5. Proprietary 2-channel CODECs have been out for ages, e.g. WMA, AVI, XHRD, SACD, DSD, DXD, Audio DVD, BlueRay Audio, HDCD etc. to name a few. MQA just adds one more.


Sent from my iPad using Tapatalk

 

[rbbert]

Again nonsense: how many times has it been stated here, that me as well as many others, are format agnostic. There are people who just want to enjoy the variety. Always these conspiracy theories where the brain is put out of service.

Vinyl, PCM, DSD, tape - anything goes.

I'm not referring to listeners; I'm referring to MQA the company, and its business relationships with the record labels. There has absolutely been an attempt by MQA to monopolize the commercial availability of "hi-res" music, which fortunately appears much less likely now than it did 2-3 years ago.

 

[rbbert]

Well, you go ahead an do statistical analysis based on a sample size of one or two. That’s what people always do when they are not able to deduct a reasoning: use anecdotal evidence.

I have no idea what you mean by this. CERN's data storage is qualitatively similar to that of any conscientious music collector; quantitatively it is significantly different. You do realize that the data error rates they discovered would result in an insignificant change (if indeed any change at all) in a digital music file? Do you know how the AccurateRip database functions?

Bottom line, master recordings in the possession of the industry are far more likely to have significant corruption than the recordings in the hands of private dedicated collectors.

 

[rbbert]

5. Proprietary 2-channel CODECs have been out for ages, e.g. WMA, AVI, XHRD, SACD, DSD, DXD, Audio DVD, BlueRay Audio, HDCD etc. to name a few. MQA just adds one more.

Most of those you mention have faded from commercial relevance. Notably SACD, DSD, DVD-Audio and Bluray Audio are "lossless" in a digital sense; not so MQA. If you still think MQA is lossless, read the nearly 100 page patent application; never are the words "digitally lossless" used together.

 

[Kuoppis]

I have no idea what you mean by this. CERN's data storage is qualitatively similar to that of any conscientious music collector; quantitatively it is significantly different. You do realize that the data error rates they discovered would result in an insignificant change (if indeed any change at all) in a digital music file? Do you know how the AccurateRip database functions?

Bottom line, master recordings in the possession of the industry are far more likely to have significant corruption than the recordings in the hands of private dedicated collectors.

Do you know what they do at CERN?

Sorry, but it’s not quite like your home computing [emoji23][emoji23][emoji23].


Sent from my iPad using Tapatalk

 

[Kuoppis]

Most of those you mention have faded from commercial relevance. Notably SACD, DSD, DVD-Audio and Bluray Audio are "lossless" in a digital sense; not so MQA. If you still think MQA is lossless, read the nearly 100 page patent application; never are the words "digitally lossless" used together.

This was a simple proof to invalidate the claim that there has not been a proprietary 2 channel CODEC before MQA.

Which of course is bollocks.


Sent from my iPad using Tapatalk