Apple iTunes - Mastered for iTunes Options

[RobertoDomenico]

Mastered for iTunes: Rolling out on a worldwide basis, Apple is now featuring songs and albums that have been specifically mastered for the iTunes Store to provide the best sound quality for the format.

More info http://www.apple.com/itunes/mastered-for-itunes/

An interesting read http://images.apple.com/itunes/mastered-fo..._for_itunes.pdf

Your thoughts?

This post has been edited by RobertoDomenico: Feb 22 2012, 18:24

[DVDdoug]

I don't see anything too bad there. The good news is that they are not promoting dynamic compression. It doesn't look like any of their "mastering tools" are even capable of dynamic compression or otherwise doing any damage (except for possible lossy file compression artifacts from the AAC format). Overall, the advice seems reasonable.

I did notice one slight error:

Because this SRC outputs a 32-bit floating-point file, it can preserve values that might otherwise fall outside of the permitted frequency range.

Of course, bit-depth and integer/floating-point have nothing to do with frequency. Change that to "amplitude range" and it's correct.

[Ron Jones]

I'm not sure why they refer to 24/96 as the ideal master, considering the final delivery SR is 44.1 kHz. The ideal should be 88.2 or 176.4, no?

[DVDdoug]

I'm not sure why they refer to 24/96 as the ideal master, considering the final delivery SR is 44.1 kHz. The ideal should be 88.2 or 176.4, no?

From what I've read, 24/96 is what most studios use. I think the point is that it's more ideal to go "directly" from 24/96 to AAC or MP3, than to convert to 44.1/16 as an intermediate step.

It turns-out that downsampling by an even number isn't an advantage, since you have to low-pass filter anyway, and filtering is more "drastic" than interpolation. If it was me, I'd probably downsample to 48kHz, rather than 44.1, but just because it "feels better"... not because there would be any audible difference.

[andy o]

This is from the pdf:

To take best advantage of our latest encoders send us the highest resolution master file possible, appropriate to the medium and the project.

An ideal master will have 24-bit 96kHz resolution. These files contain more detail from which our encoders can create more accurate encodes. However, any resolution above 16-bit 44.1kHz, including sample rates of 48kHz, 88.2kHz, 96kHz, and 192kHz, will benefit from our encoding process.

How is this true if it's gonna be downsampled to 44.1/16 anyway?

[db1989]

I assumed they’re trying to push the ‘AAC has no inherent bit-depth’ line there.

[skamp]

^ Apparently the Nero AAC encoder takes 24bit/96kHz files as input, no problem, resulting in 96kHz encodes… Get ready for 512kbps high-res AAC tracks on iTunes?

This post has been edited by skamp: Feb 22 2012, 22:05

[punkrockdude]

I have tried out a couple of encoders how they handled audio below 16 bits and the audio was around -120 dBFS. LAME, Quicktime AAC and Nero AAC Encoder encoded the signal as clear as if it would have been up around 0 dBFS. Vorbis had sound but more of a low frequency transient response when drum hits were sounding and not exactly much high frequency. Fraunhofer AAC encoder actually managed to somehow apply gain to the signal so that it went up to somewhere around -60 or -70 dBFS instead of -120 dBFS. I used both low and high bitrates with each encoder. I did the test about a month ago so if you proof then download the versions of the encoders that were up to date around then. I should mention that I used 44.1kHz/24 bit PCM.

This post has been edited by punkrockdude: Feb 22 2012, 22:15

[andy o]

I assumed they’re trying to push the ‘AAC has no inherent bit-depth’ line there.

^ Apparently the Nero AAC encoder takes 24bit/96kHz files as input, no problem, resulting in 96kHz encodes… Get ready for 512kbps high-res AAC tracks on iTunes?

But doesn't any psychoacoustics based codec work with stuff we can hear? I don't know the specifics of AAC but it should cut the frequencies above 20kHz at least, shouldn't it?

[C.R.Helmrich]

Fraunhofer AAC encoder actually managed to somehow apply gain to the signal so that it went up to somewhere around -60 or -70 dBFS instead of -120 dBFS.

That shouldn't happen. Could you clarify how you created your test file? And I assume you used Winamp for encoding?

Indeed, any decent encoder should also accept 24-bit input and resample prior to encoding if necessary. So nothing iTunes-exclusive here.

andy o: yup, makes no sense really, but could be called "high-res AAC".

Chris

P.S.: check out today's doodle on google.com: Happy 155th birthday, Heinrich "Hi-Res" Hertz!


This post has been edited by C.R.Helmrich: Feb 22 2012, 23:08

[punkrockdude]

Fraunhofer AAC encoder actually managed to somehow apply gain to the signal so that it went up to somewhere around -60 or -70 dBFS instead of -120 dBFS.

That shouldn't happen. Could you clarify how you created your test file? And I assume you used Winamp for encoding?

Indeed, any decent encoder should also accept 24-bit input and resample prior to encoding if necessary. So nothing iTunes-exclusive here.

I have Ubuntu on this computer but I will install Winamp, create a beat, lower the volume and encode it and get back to you. Damn, I just started watching a new series ;D Regards.

[andy o]

personally, when they stop beating around the bush and offer lossless already, I'll finally buy my first song or album digitally.

[mixminus1]

I had to read through the PDF a couple times to make sure I stifled some knee-jerk reactions, and overall, it's pretty reasonable, and actually quite informative (to the lossy-encoding noob) at times.

@db1989: In reference to your response about AAC's internal bit depth, they more-or-less do say it a couple times:

It is this 32-bit floating [-point] file that’s used as the input to the encoder...

Regardless of the bit depth of the original source file (16- or 24-bit), you should generate a 24-bit file to preserve maximum fidelity resulting from the AAC coding process.

It's also nice to see the afclip utility and its tally of inter-sample overs.

I think they still needlessly cast 16/44 audio in a negative light, but at least they provide mostly technically-sound reasons for their request for 24/96 files (no dithering needed, for instance).

@andy o: Sadly, a lossless file does not necessarily equal the original audio - this was from August of last year:

http://www.hydrogenaudio.org/forums/index....showtopic=89818

Clearly-audible artifacts (watermarking) on FLAC files sold by an online retailer (Passionato) that were completely absent from the same tracks purchased from Deutsche Grammophon's online store. I've since heard similar artifacts on an album that I downloaded from iTunes and that was distributed by UMG (Florence + the Machine's "Ceremonials") - once again, completely absent from a verified FLAC rip.

[db1989]

[…] overall, it's pretty reasonable, and actually quite informative (to the lossy-encoding noob) at times.

Although I only skim-read the first half or so, this is basically what I thought. It’s a total PR piece, obviously; but it’s somewhat redeemed by the fact that they cast a critical eye over the loudness war, explicitly warn against pointlessly upconverting, and a few other things like this—which could do with the extra publicity gained from a mention by such an influential company.

@db1989: In reference to your response about AAC's internal bit depth, they more-or-less do say it a couple times: […]

Thanks! But I did already see those; that’s what I meant by my comment. Assumption wasn’t the best choice of word!

I think they still needlessly cast 16/44 audio in a negative light, but at least they provide mostly technically-sound reasons for their request for 24/96 files (no dithering needed, for instance).

As with the first quote/paragraph above, I agree.

[punkrockdude]

Winamp didn't want to work for me under WINE but I used the latest dll with the wrapper and encoded through Foobar2000. It did not get the results I expected. I want to remember that when I encoded with Winamp the sound somehow was normalized up to -60 or -70 dBFS but this time that did not happen. It is still very muddy sounding though. All the hihats, snares and high frequencies are gone and left is just dull kick drum hits.

This post has been edited by punkrockdude: Feb 23 2012, 00:57

[punkrockdude]

Helmrich:

Well the issue with the sound being louder using Fhg encoder didn't seem to be a problem with the exe wrapper but here is how different encoders sound on a source where the hight peak hit -96dBFS and then normalized so that we can hear it. The wav file is the source file I used to encode all samples with. All encoders used cbr 320 kbps.

http://speedy.sh/d6SBA/Untitled-Song-v3-96dbfs-24bit.wav
http://speedy.sh/EmCrK/Untitled-Song-v3-96...normalized.flac
http://speedy.sh/MfY3V/Untitled-Song-v3-96...normalized.flac
http://speedy.sh/H7aDd/Untitled-Song-v3-96...normalized.flac

This post has been edited by punkrockdude: Feb 23 2012, 01:38

[RobertoDomenico]

Doesn't look like we'll be seeing ALAC now for some time if ever, a shame.

[icstm]

From the pdf:

Movies, for
example, have very detailed standards for the final mastering volume of a film’s
soundtrack.

what are these and could they apply to the music industry?

[Brand]

I think movies have standards for both audio and video, so that you get the ~same experience in every theater. Music usually isn't consumed in controlled environments like that, so it makes less sense to apply some official standards.

[ozmosis82]

Did anyone else notice how the PDF also mentioned that Sound Check can be applied across albums (e.g DSOTM)? This is news to me.

[C.R.Helmrich]

Well the issue with the sound being louder using Fhg encoder didn't seem to be a problem with the exe wrapper but here is how different encoders sound on a source where the hight peak hit -96dBFS and then normalized so that we can hear it. The wav file is the source file I used to encode all samples with. All encoders used cbr 320 kbps.
...

Thanks a lot for this investigation! The AAC encoder will preserve more than the bass drum hits in the next version of Winamp.

And sorry for hijacking this thread. So, back on the subject: In the PDF it says,

As technology advances and bandwidth, storage, battery life, and processor power increase, keeping the highest quality masters available in our systems allows for full advantage of future improvements to your music.

Does that mean that, when you upload your hi-res masters, Apple actually keep them on their servers? I wonder if every uploader will be fine with that. Maybe they are seriously considering offering better-than-AAC256 encodings in the future, i.e. more bit rate and sampling rate? Interesting.

Chris

This post has been edited by C.R.Helmrich: Feb 23 2012, 21:01

[IgorC]

About AAC and 24/96
http://www.iis.fraunhofer.de/Images/AES547...cm182-51651.PDF

The AAC algorithm performs the t/f mapping by means of an MDCT with a maximum resolution of 1024 spectral lines. A typical fast and efficient implementation can be realized by a 10 stage radix 2 butterfly operation. Here each butterfly stage reduces the available precision by 0.5 bit in average. Therefore, this implementation loses about 5 bits precision inside the MDCT transformation. In case of a 24 bit data representation, a final precision of only 19 bit can be achieved after the t/f mapping. For most of the remaining parts of the encoder it is in general sucient to be able to handle the dynamic range of 24 bits.

19 bits is very close to state-of-art DACs (effective 20-21 bits). Probably only few percents of people have hardware that will expose the benefit of 19 or more bits. Even less who would actually be able to hear it.

This post has been edited by IgorC: Feb 23 2012, 21:19

[JunkieXL]

True, but it is still a selling point for their product. I doubt that the majority of iTunes users actually read into this sort of thing. It's like a new shiny design or add on that helps people feel better about their purchase.
JXL

This post has been edited by JunkieXL: Feb 23 2012, 22:04

[punkrockdude]

Thanks a lot for this investigation! The AAC encoder will preserve more than the bass drum hits in the next version of Winamp.

No problem at all. I like doing all kinds of test so if you want a beta tester then I am in. Regards.

[Kujibo]

This was a bit of a frustrating read:

http://arstechnica.com/apple/news/2012/02/...he-ipod-age.ars

However, I was quite encouraged by the majority of the user comments. I'd have to think the followers of ars technica are more technically sophisticated than your average site's followers.

I do however hope that Apple is trying to put a system in place to accept high quality masters in the hope that they move to providing lossless at some point. Short of actually providing lossless this whole mastered for iTunes thing seems like yet another way to market a new way to sell the same music, and sadly I don't even see a concrete jump in technology here, I don't really see what it is guaranteeing at all.