Apple Music Launches 'Digital Masters' Initiative Out of Old 'Mastered for iTunes' Program

[B/D]

Fair enough, thank you very much for the detailed explanation. But then I wonder, it´s possible that things develop the other way around?.

That is, if there is no much room for improvements in lossless audio encoding, maybe there is in lossy encoding. Maybe a ultra small, high fidelity format which is transparent to the original can be developed, thus reducing even more infrastructure and peering costs. Much more small and efficient than AAC and Opus today. Would that be possible?

 

[Rigby]

That is, if there is no much room for improvements in lossless audio encoding, maybe there is in lossy encoding. Maybe a ultra small, high fidelity format which is transparent to the original can be developed, thus reducing even more infrastructure and peering costs. Much more small and efficient than AAC and Opus today. Would that be possible?

Lossy perceptual codecs have already improved by leaps and bounds over the last two decades or so (which is one reason why lossless digital music never really got a foothold in the market). It's also not just a matter of new formats, but how a specific encoder implementation uses the tools provided by the format. Just look at how much MP3 encoders have improved from the early Fraunhofer implementations to e.g. recent versions of LAME. It's day and night. Modern perceptual codecs at bitrates >=256 kbps or so are very close to lossless most of the time (there are some sounds where you can still hear differences if you know what to listen for).

 

[B/D]

I agree, and my question was... could they (lossy codecs) improve even more in the future?

 

[Rigby]

I agree, and my question was... could they (lossy codecs) improve even more in the future?

I'm not really following the latest research closely anymore, but I don't expect big leaps anymore when it comes to high-end perceptual encoding. The focus in groups like MPEG today is more in encoders that use very little bandwidth but still deliver acceptable quality (HE-AAC, DD+ low-bitrate mode etc.) and immersive 3D audio formats.

 

[xyz01]

FLAC and ALAC are lossless, yes, but lossless to a CD quality 44khz, 16bit file. ntlman was referring to purchasing 24bit high resolution files from a place like HDTracks which are distributed in the AIFF/WAV formats. From a comparative standpoint, a 24bit/192 file in AIFF/WAV is much better than a FLAC/ALAC of a 16/44 source - also way larger.

That was not what he wrote: " I'm not even sure why in 2019 anyone bothers with FLAC or ACC instead of just the full size AIFF/WAVs when storage is cheaper than ever."

Just to be clear, for a given sample rate and bit depth ALAC/FLAC and AIFF/WAV have identical content - ALAC/FLAC are just smaller. HDTracks sells FLACs, I know I've bought some there in the past.

That said, even on good equipment it's hard hearing the difference between 44.1/16 and 96/24. I could hear it on some jazz recordings, but I'm inclined to think that has more to do with different masters.

What has stood out for me in sound quality has been SACD, but I'm sure that has more to do with separation of channels and mastering as well.

 

[ntlman]

That was not what he wrote: " I'm not even sure why in 2019 anyone bothers with FLAC or ACC instead of just the full size AIFF/WAVs when storage is cheaper than ever."

Just to be clear, for a given sample rate and bit depth ALAC/FLAC and AIFF/WAV have identical content - ALAC/FLAC are just smaller. HDTracks sells FLACs, I know I've bought some there in the past.

That said, even on good equipment it's hard hearing the difference between 44.1/16 and 96/24. I could hear it on some jazz recordings, but I'm inclined to think that has more to do with different masters.

What has stood out for me in sound quality has been SACD, but I'm sure that has more to do with separation of channels and mastering as well.

Hey @jeffkoontz and @xyz01: thanks for responding.

You're actually both right, in terms of the points I was trying to make. For Mac users, you need a separate app other than iTunes to play FLAC files (I'm one of those weirdos who actually thinks the iTunes desktop app is perfectly fine). You can easily set up via Apple's audio midi setup to output at the highest available frequency through your connected DAC. My iPhone and iPad are a different story: I have to use the ONKYO HF audio app, because the iOS' iTunes does not natively drive and configure an external DAC output setting (I'm wondering if this will change in iOS 13 on the new Music app). The ONKYO app lets me play on iOS 96/24 WAVs I've bought of Radiohead and Beatles albums and enjoy them as fully as I could on macOS. It also allows me to upsample files all the way to 384/32, which sounds really nice.

Not every album is available for sale in FLAC format, but every album still comes out on CD. I live in LA, so I have access to a great record store (Amoeba), as well as an incredibly well stocked public library system. My point was: why bother ripping the files from the CD and converting them to FLAC (which requires a separate app not included with macOS), and then find another player utility/app other than iTunes to play them back? Especially when I no longer need to worry about saving storage space?

As for Apple's Lossless format, my own take is not scientific, but based on my own personal observations in my audio setup. I had been using ALAC for years, and then one day decided to compare to an AIFF rip of the same file. To my ears, I could notice a significant improvement over the ALAC, and have simply ripped full size AIFF files ever since.

I agree, SACD is great: I've picked up Bjork's "Post" and NIN's "With Teeth". Even compared to the remastered standard PCM files included on the discs, the 5.1 surround remastering blew me away.

 

[xyz01]

Hey […] @xyz01: thanks for responding.
[…]
As for Apple's Lossless format, my own take is not scientific, but based on my own personal observations in my audio setup. I had been using ALAC for years, and then one day decided to compare to an AIFF rip of the same file. To my ears, I could notice a significant improvement over the ALAC, and have simply ripped full size AIFF files ever since.
[…]

ALAC is identical in quality to the AIFF (and FLAC) - you can convert the AIFF to ALAC and back again to AIFF without losing a bit of sound content. I really doubt you hear a difference, as there isn't any. Of course, there could be a bug in the decoder, but that sounds extremely unlikely.

 

[Kram Sacul]

Proper mastering makes a far bigger difference than the minute (and probably imperceptible) differences between 256kbps AAC and lossless encoding. One of the biggest benefits of the "Mastered for iTunes" program in my book is that the songs usually use less dynamic range compression, which prevents distortions and other problems (google "loudness wars" if you want to know more). That often makes them sound better than the corresponding CDs.

Bingo. People say they want lossless but if the sound file is brickwalled to death with no dynamic range who cares? it’s going to sound like junk no matter what format.

 

[ntlman]

Yeah, I'm all over the place when it comes to mastering. Generally speaking, I don't think any modern rock albums with professional distribution (ie indies and major labels) released after 1995 need a re-master unless something went horribly wrong at the mastering studio or CD factory. That said, I love all the Rolling Stones re-masters from the mid 90s and early 00s (the gold standard, IMHO, for re-master efforts) and only 50/50 on the U2 re-masters from the 00s through today. The ones remastered by Scott Sedillo are subpar, and are the ones used with the MFiT standard.

I went from having the original 1980s CD issue of the Beatles Sgt. Pepper to the 50th anniversary version and thought it was great. I originally attributed it simply to the re-master... until I heard the 2009 remasters of the Beatles catalog. Those were definitely improved, and I realized the 2017 Sgt Pepper re-issue benefitted more from the "re-mixing" that also was applied (which, controversy aside, I thought was tasteful and really well-done).

To me, the issue is that MFiT seems to sap some of the warm bottom end, IMHO. Giles Martin (Sgt Pepper remaster engineer/son of Sir George) himself has said that having worked on MFiT, he did not think it was at the level that it should be, in particular for streaming. The U2 remasters Sedillo supervised are some of the more bottom-end heavy albums in the U2 catalog, and when his remasters are compared to files ripped from the original issued CDs, the criticism seems to bear out.

 

[xyz01]

Bingo. People say they want lossless but if the sound file is brickwalled to death with no dynamic range who cares? it’s going to sound like junk no matter what format.

I guess those of us who want lossless aren't exclusively listening to music that's massively compressed.

 

[Rigby]

To me, the issue is that MFiT seems to sap some of the warm bottom end, IMHO. Giles Martin (Sgt Pepper remaster engineer/son of Sir George) himself has said that having worked on MFiT, he did not think it was at the level that it should be, in particular for streaming. The U2 remasters Sedillo supervised are some of the more bottom-end heavy albums in the U2 catalog, and when his remasters are compared to files ripped from the original issued CDs, the criticism seems to bear out.

Well, this seems to be a bit misleading. "Mastering for iTunes" is not something that Apple does or controls. It's essentially a collection of guidelines for audio engineers, along with a set of software tools that allows the engineer to test how their master will sound in AAC format before submitting it to Apple (most importantly that includes an AAC encoder toolchain that produces the exact same results as the one that Apple uses when encoding files for the iTunes store). However, actually creating the master is still the engineer's responsibility, so if something is wrong with it they have nobody to blame but themselves.

 

[B/D]

Actually, Apple do control it to some extent. If clipping or brickwalling are present in an audible way, then the album is "rejected" and it doesn´t feature the MFI badge in the Store. The same also happens if the master used to create the final for sale AAC files isn´t "high resolution" enough.

 

[HDFan]

If headphone sound quality is of interest you might want to check out true-fi. The improvement in my mediocre Beats Studio 3 listening experience was amazing:

https://www.sonarworks.com/truefi

They have a listening configuration test where you can determine your specific sound profile.

https://www.soundandvision.com/content/true-fi-sound*************app-now-supports-200-headphones

there has yet to be a study where people can prove it in a blind test.

Here is a peer reviewed paper, based on 80 studies, which concludes that people can hear the difference:

https://www.audiostream.com/content/its-official-people-can-hear-high-res

 

[B/D]

MP3 and AAC encoders typically don't have enough headroom to handle the very high peak level of complex and nuanced recordings , and so introduce extra clipping distortion as well as all the encoding artefacts. This is what the Mastered for Itunes initiative, now rebranded "Digital Masters", have been trying to avoid, combining the guidelines and tools provided by Apple with their state of the art, and very much improved trough the years, AAC encoder.

Again, all the technicals details can be found here. Is not just marketing, it´s a good thing. By the way, that PDF have been just updated, and it now says this at the beginning

Note: Apple Digital Masters replaces Mastered for iTunes as the name of the program to better reflect the fact that these audio advantages are available to our entire music catalog across the Apple ecosystem, whether streamed or downloaded. All Mastered for iTunes releases are now badged as Apple Digital Masters. All of the Mastered for iTunes software tools are still usable for Apple Digital Masters creation.

 

[Rigby]

Actually, Apple do control it to some extent. If clipping or brickwalling are present in an audible way, then the album is "rejected" and it doesn´t feature the MFI badge in the Store. The same also happens if the master used to create the final for sale AAC files isn´t "high resolution" enough.

Yes, but you know what I meant. They are not involved in creating the master, hence responsibility for any flaws lies with the audio engineer.

 

[Diatribe]

They are stored in a 32-bit floating point value, so you're still getting the full 24-bit audio... although at 256. Compare this to a 256 rip a couple of years back... night and day.

 

[eas]

but listened to via $15 earbuds through the iPhone DAC.
Hi-Fi on an iPhone is an oxymoron

$15 earbuds are inherently garbage. The DAC is fine.
[doublepost=1565720927][/doublepost]

[doublepost=1565197024][/doublepost]

Cool! But I guess it won´t be live soon, since current Airpods only support AAC.

Firmware updates are a thing that happens.

 

[B/D]

A firmware update won´t make the airpods support a new high resolution lossless audio codec. The W1 and H1 chips only support ( via hardware decoding, in fact, for efficiency reasons), AAC. That´s it. Software decoding wouldn´t be a good idea, since we´re talking about truly wireless earbuds, where battery life is critical.

 

[smileman]

Deezer?

In my case also needs to be integrated into Roon
[doublepost=1566887599][/doublepost]

How is Tidal HD getting from your watch to your ears? (Answer AA


How is the watch delivering the music to your headphones? If the answer is bluetooth, your $20 a month Tidal HD stream is getting converted to a 256bitrate AAC codec in the process.

Yes, but I use Tidal in other ways where I can take full advantage of lossles and MQA, such as via Roon or through a wired connection from my iPhone eg Dragonfly USB DAC.

But you’re right, there will be no way to take full of vantage of hi-fi Tidal via the watch that I can think of.

 

[ZeitGeist]

In other news, Apple continues to invest in relatively fringe and useless features, yet continues to not fix glaring bugs and flaws in Apple Music.