Does "Mastered for iTunes" matter to music? Ars puts it to the test

[wgscott]

Link to Ars Technical article

 

I am in no way affiliated with Arse or Mr. Wednesday, or Guy Fawkes.

 

 

We came away from the process learning that it absolutely is possible to improve the quality of compressed iTunes Plus tracks with a little bit of work, that Apple's improved compression process does result in a better sound, and that 24/96 files aren't a good format for consumers.

[AudioDoctor]

Link to Ars Technical article

 

I am in no way affiliated with Arse or Mr. Wednesday, or Guy Fawkes.

 

I however am about to go all Guy Fawkes on them and blow up their server with my emails telling them to stop publishing such false and misleading conclusions.

[Akapod]

From Ars:

 

The fact is that Apple is now the number one retailer of music in the US—meaning a fairly large percentage of music today is purchased in 256kbps iTunes Plus format. Though our hard drives have grown into multi-terabyte monsters, compressed audio remains the most common music delivery format.... At some point, we'll all be downloading uncompressed audio over widely available LTE Advanced wireless connections to our smartphones with well more than 64GB of storage. But that day is not today—and iTunes Plus could foreseeably be around for the rest of this decade..

 

Apple already has a system which provides for downloading "standard" files to portable devices, while allowing for the downloading of hi-res files to a desktop computer. That's how video works in iTunes. Would love to see them use the same system for audio.

 

(And, oddly, HD videos are only available with Dolby Digital files, not with Dolby TrueHD. So we're getting MP3-quality sound along with near-Blu-Ray quality video.)

[bleedink]

Originally Posted by Arse

We came away from the process learning that it absolutely is possible to improve the quality of compressed iTunes Plus tracks with a little bit of work, that Apple's improved compression process does result in a better sound, and that 24/96 files aren't a good format for consumers.

 

 

THIS! Then they provide absolutely no justification for this other than our electronics and ears can't keep up. Scroll down a few paragraphs and the author is in the recording studio listening to an actual master which he says sounds "airy", contrasting it with the "boxy" sounding compression. OK so which is it? Then he goes on to keep saying how he can tell the difference between various compression schemes. Really? It would seem that studio master doesn't sound so bad...

[bleedink]

Sorry to go on a diatribe but why doesn't iTunes do it like HDtracks? Put the 16/48 file on analog tape and digitize it at 24/192 then master it for iTunes? Wouldn't that sound even better at 256kbps?

[souptin]

Here's my favorite quote:

 

"Moving to 24 bits increases the dynamic range to 144dB. Assume that you have audio equipment capable of reproducing the entire 144dB range of sound—and we're not aware of any equipment available to consumers which can do so—and that you have turned it up loud enough to hear that full range. The loudest sounds would be loud enough to cause permanent hearing loss or possibly even death."

 

There I was thinking that the "Mastered for iTunes" was all about trying to squeeze some more money out of yet another release of the same music, but in actual fact Apple is trying to save us from certain death caused by loud music. Awesome!

[The Computer Audiophile]

Hi Guys - I thought it was a pretty good article despite the fact I disagree with some of the things said. Chris Foresman does a pretty good job writing in depth about iTunes audio.

 

iTunes HD Audio at 24 bit / 48 kHz was supposed to happen and now seems derailed. That's a bummer. Hopefully Mastered for iTunes is just an intermediate step on the path to iTunes HD Audio.

[AudioDoctor]

Hi Guys - I thought it was a pretty good article despite the fact I disagree with some of the things said. Chris Foresman does a pretty good job writing in depth about iTunes audio.

 

iTunes HD Audio at 24 bit / 48 kHz was supposed to happen and now seems derailed. That's a bummer. Hopefully Mastered for iTunes is just an intermediate step on the path to iTunes HD Audio.

 

I would happily settle for 16/44 ALAC files.

[The Computer Audiophile]

I would happily settle for 16/44 ALAC files.

 

 

As would I.

 

If we could purchase 16/44.1 lossless from iTunes the local record shops would stop selling music and only sell pipes and pop culture trinkets. Oh wait, that's nearly happened already.

[AudioDoctor]

As would I.

 

If we could purchase 16/44.1 lossless from iTunes the local record shops would stop selling music and only sell pipes and pop culture trinkets. Oh wait, that's nearly happened already.

 

 

subtle Electric Fetus reference only AudioDoctor will get... ;-) I go there a lot and buy CDs and a lot of Vinyl, but I have to walk through all the crap to get to it. They have to do it though to stay afloat.

 

and yes, I of course meant 16/44 lossless files.

[Richard Dale]

Hi Guys - I thought it was a pretty good article despite the fact I disagree with some of the things said. Chris Foresman does a pretty good job writing in depth about iTunes audio.

 

iTunes HD Audio at 24 bit / 48 kHz was supposed to happen and now seems derailed. That's a bummer. Hopefully Mastered for iTunes is just an intermediate step on the path to iTunes HD Audio.

 

Oh I'm sorry to hear about 24/48 not happening. I think you have hinted before that something 'hi res' was going at Apple but couldn't say anything while it was still a going concern. And now it is dead, you can - oh dear! I thought the Ars Technica article was well written and a good read. I'm very keen on any mainstream articles about sound quality issues with music formats. Even if I would rather we were talking about 24/48 or 24/96 music for iTunes, but at least it is a start when people are talking about improving AAC.

 

What I find depressing is all the 'objectivists' who come out and attack anyone who thinks they can hear hear the difference between AAC and WAV, or between 16/44.1 and 24/96. And all the nasty attacks on 'audiophiles' (look for 'Monster Cable' or other code words when that is happening on the Ars Technica forum discussions).

 

Even mastering engineers are not immune, and they are not supposed to be able to distinguish between the different formats and conversion rates without double blind tests. How could they do their job? I wrote a post pointing that out. I want hi res formats to be easily available and mass market, just like CDs (and not like SACDs). But how can we get started with an advocacy compaign when we are portrayed as 'natters' by the objectivists?

 

I know we have some people on the forum on this site who favour double blind tests, but they are not so fanatical, and at least we can agree to differ (typically after about 150 posts!). I think pretty much everyone thinks DBTs are a perfectly useful tool, it's just that not everyone thinks they are the only useful tool, or a substitute for trained listeners. So when a guy was saying the article was so bad it should be taken down because the mastering engineers weren't using DBTs, I got annoyed. Should every article on audio on mainstream sites need to go through Hydrogenaudio for approval before publication? I think that would be very sad. I have wondered if we share the same hobby as those guys, and I think my current opinion is that we don't.

[wgscott]

I'm a big fan of Ars Technica and tend to take their stuff a lot more seriously than most sites on the inter-web. When they cover subjects that I know something about, or in a couple of cases even something I did, they seem to get it right far more frequently than pretty much any other popular news source out there.

[Jud]

I'm a big fan of Ars Technica and tend to take their stuff a lot more seriously than most sites on the inter-web. When they cover subjects that I know something about, or in a couple of cases even something I did, they seem to get it right far more frequently than pretty much any other popular news source out there.

 

Yeah, generally agreed. That's one reason I was rather distressed to read the following:

 

Human hearing, mathematics, and audio equipment all have their limits. Combined, they make selling direct-from-the-studio, 24-bit 96kHz (some files come in sampling rates as high as 192kHz) audio files impractical, if not outright useless.

 

The link is to a Chris Montgomery article discussed here at some length re hi-res files not resulting in any difference in sound quality vs. Redbook. I corresponded with the article's author, and I think it is fair of me to say he relies quite heavily on the Boston Audio Society / AES paper regarding the inability of listeners to identify hi-res from Redbook in blind A/B testing. I would regard such conclusions, especially given the differing conclusions of other well-respected audio professionals, as provisional, and wish Ars would have gone a little less "all in" for that view.

 

Ars in fact goes even further by liberally quoting an audio engineer to the effect that lossy codecs don't differ to any appreciable extent in audio quality from Redbook, and that he has the "null tests" to prove it.

 

So there you go - Lossy Sound Forever! Perfect!

[Blu]

The article brings up points I'm sure have been already aired at CA. Such as 1) 16 bit files give a dynamic range of 96dB, 2) 24 bit files, 144 dB dynamic range and that 3) modern recordings don't have a dynamic range beyond 60dB. Which begs the question, why go to 24 bit?

 

Hearing range for most young humans is 20 - 20,000 Hz, which reduces with age. While the frequency range from 16/44.1 files (CD) goes up to 20,000 Hz. Most amps are built to cover near the frequency range of 20 - 20,000 Hz. Now if these amps are fed frequencies higher and lower than that range, because of non linearities these frequencies can be shifted into the audible range as intermodulation distortion.

 

Do more knowledgeable people here, question the validity of these points?

 

Secondly, do you find the music from higher bit / frequency music files sound more real, provided it is a good recording to start with?

 

Blu

[Jud]

The article brings up points I'm sure have been already aired at CA. Such as 1) 16 bit files give a dynamic range of 96dB, 2) 24 bit files, 144 dB dynamic range and that 3) modern recordings don't have a dynamic range beyond 60dB. Which begs the question, why go to 24 bit?

 

Hearing range for most young humans is 20 - 20,000 Hz, which reduces with age. While the frequency range from 16/44.1 files (CD) goes up to 20,000 Hz. Most amps are built to cover near the frequency range of 20 - 20,000 Hz. Now if these amps are fed frequencies higher and lower than that range, because of non linearities these frequencies can be shifted into the audible range as intermodulation distortion.

 

Do more knowledgeable people here, question the validity of these points?

 

Secondly, do you find the music from higher bit / frequency music files sound more real, provided it is a good recording to start with?

 

Blu

 

Don't listen to me. Rather, read what a couple of very smart guys say about it in the context of a paper for the Audio Engineering Society: http://www.goodwinshighend.com/music/hdcd/aes_paper.pdf

 

It's a 26-page paper that will give you some idea of the extreme care and thoroughness with which the various issues bearing on the subject were considered. If you'd like something less technical, I can give you links to a couple of interviews that sum things up to a certain extent.

 

I won't pretend this particular paper, which doesn't even specifically treat the subject of hi-res, settles the subject. But it does impress me as a good, careful piece of work.

[Garf]

I have to disagree with the statement "3) modern recordings don't have a dynamic range beyond 60dB. "

 

Most recordings perhaps, not all, and certainly not classical music. The article pretty much says that CD always sounded better regardless. It is in no way surprising that they can make 256kbps lossy itunes crap sound better. But no matter how many coats of paint you put on it, a turd is still a turd.

 

G

[Julf]

If we could purchase 16/44.1 lossless from iTunes the local record shops would stop selling music and only sell pipes and pop culture trinkets. Oh wait, that's nearly happened already.

 

And meanwhile we need Apple to tell record labels and producers to avoid clipping...

[Blu]

Don't listen to me. Rather, read what a couple of very smart guys say about it in the context of a paper for the Audio Engineering Society: http://www.goodwinshighend.com/music/hdcd/aes_paper.pdf

 

It's a 26-page paper that will give you some idea of the extreme care and thoroughness with which the various issues bearing on the subject were considered. If you'd like something less technical, I can give you links to a couple of interviews that sum things up to a certain extent.

 

Thank you Jud, some quick skimming suggests that it is best to download the music file in the format it was recorded in, as problems can be introduced with downsampling.

 

When I have more time I will read through the full article, especially since it is by Keith Johnson, I would be interested in reading the other articles you have suggested.

 

Blu

[Jud]

When I have more time I will read through the full article, especially since it is by Keith Johnson, I would be interested in reading the other articles you have suggested.

 

Blu

 

Be glad to after I have a chance to do some searching. May be a little while. :-)

[bliss53]

It seems that the music industry has been "mastered for itunes" for years. See the "loudness wars" at http://www.youtube.com/watch?v=3Gmex_4hreQ. It is only now that Apple is acknowledge higher resolution files with uncompressed mastering. But they say we are not ready for it.