Mastered For iTunes


Over the last year, many of you have probably seen the term “Mastered For iTunes” start popping up on the iTunes store for popular new releases.  I’ve had a lot of people asking me what this means, and if they need it for their masters, so it’s time to talk about this new initiative from Apple.

Now, before I go any further, I just want to say that I know how anything “Apple” can be a very polarizing thing, people either love them or hate them and want nothing to do with iTunes and like.  Before you stop reading if you’re one of the later, just hang in there a little bit longer and I think you’ll appreciate why Mastered For iTunes (MFiT) is a good thing for ALL fans of music.

So then, what is MFiT and what is the difference between it and normal iTunes songs?  The short answer is that for the last 4-5 years, Apple has been working with some of the best mastering and recording engineers in the world, as well as the Producers and Engineers branch of the Recording Academy, to improve the sound quality of what gets sold on the iTunes store.  This is done through two main avenues:

1. Allowing artists/labels to submit higher resolution formats to the Apple store, instead of the CD standard 16bit 44.1kHz format that was required before.  Using a higher resolution allows for much more transparent AAC conversions, as well as providing a more future proof version of all the songs they sell.  For instance, if one day Apple decides to sell high-resolution “audiophile” versions of songs, they already have those in their archives if they were submitted in MFiT format.

2. The second aspect of the MFiT initiative was collaborating with some of the best audio engineers in the music industry (i.e. Bob Katz, Bob Clearmountain, and Bob Ludwig to name but a few) to improve the AAC encoding process to make it as transparent as possible.  This has a benefit to all music lovers, since everything iTunes sells or any AAC encodes you make yourself in iTunes benefit from the new AAC improvements.  For instance an AAC you made in the last 12-14 months will likely sound a lot better than one made 4 years ago.

I’ll talk about this some more at the end of this article.

In addition, Apple has some brand new tools they’ve made available to anyone, designed to let you hear exactly how your song will sound when they convert it to AAC before it goes on the iTunes store (or before you convert it using iTunes yourself, say for a new CD).  This is done using the new AURoundTripAAC Audio Unit plug in, which you can download here:

You’ll also find the full technical details of the MFiT program there if you want more specifics than I have space to go into here.

So then, how does this all work from the standpoint of the producer?  Basically instead of creating a CD ready 16bit/44.1kHz wav file when mastering, you’re instead going to be creating higher resolution formats to supply to the Apple Store.  Files can be 24 or 32bit, and sample rates up to 96kHz are supported.  You can either let Apple convert the sample rate down with their SR convertor (said to be mastering quality by Bob Katz), or do the SR conversion yourself if you prefer having that control, and then submit the tracks at 44.1Khz, it’s up to you.


In addition, you can insert the AURoundTripAAC plug in at the end of your mastering chain, and see exactly how any mastering processing you’re doing could lead to peak or intersample clipping in the AAC convertor Apple uses.  This lets you tailor the mastering specifically for the AAC encoder, to ensure the best sounding downloads from the Apple store.  Typically (and not surprisingly), this leads to songs being slightly less loud than we are probably used to, which is not really a bad thing IMO.

But the AAC Round Trip plug in has one more trick up it’s sleeve, and that’s a built in blind listening test so you can see just how close to the source file you can get your AAC encodes.  While playing back your song, the plug in will randomly play the processed audio and the source audio, and you need to guess which is the source.  When you’re done, it will tell you how many of the guesses out of 20 you got right.

If nothing else this is a pretty impressive way to see just how far AAC encoding has come.  In the studio I can usually guess the source about 95% of the time, but it’s not easy at all and takes me quite awhile to find a point in the song where it’s apparent.  In contrast, I can spot 320kbps MP3’s compared to the source wav reliably within a few seconds usually.


So, this is all great news for musicians, right?  Well, it will be in time, but for now unfortunately it seems that mainly the major labels are able to submit MFiT files to the Apple Store.  Some aggregators like CD Baby can submit MFiT files for you, but the process is slow and still more costly than a normal submission.  At the very least there’s no reason artists can’t prepare MFiT compatible versions of their songs now, and resubmit those when the option becomes available or cheaper.

In the meantime, at least we all benefit from having better quality iTunes versions of our songs thanks to the new AAC encoder improvements.  And it’s nice to know that Apple is taking audio quality so seriously that they are reaching out to the best mastering engineers in the world soliciting ways to improve what they offer.  Not to mention getting ready to have higher resolution versions of our songs on file for any possible future improvements.  Making things sound better is rarely a bad thing in my book!

On that note, a lot of people have been asking me, do 256kbps AACs really sound better than 320kbps MP3s?  The answer rather frustratingly is, that depends.

When I learned that Apple had been improving the AAC encoder over the last year with the help of some of my mastering idols, I was of course intrigued and sat down to do my own detailed listening tests across a wide range of material (Dubstep to Debussy).  What I found was that newer AACs definitely sounded a lot better than the few I ripped years ago, it was pretty noticeable in fact.  This was good news, except for the fact that I have always been converting my wavs and CDs to 320kbps MP3s (typically with the LAME convertor, sometimes with Fraunhofer) over the last 10 years.  Was it really worth it to re-encode everything as AAC, and did it sound better than my MP3’s?

Here’s where things got interesting.

When comparing AAC and MP3’s to the source material, AAC pretty much always won as sounding the most like the source material.  Actually, for me it did always win.  But when comparing AAC to MP3 to see which I thought sounded better in a blind test, I choose each about 50% of the time.  The differences are subtle, but on some tracks the 320kbps MP3s had a slight edge to the sound, a brightness that I preferred even though the AAC version sounded closer to the original.

Still, I’m a freak about audio quality, so I have a feeling I’m going to be spending some time re-encoding my MP3 collection to AAC since it’s closer to the source audio.  I’ll probably only bother to do this for my favorite albums, and things I listen to often, but I think the improvement in audio quality is worth the time it takes (and honestly iTunes encodes are pretty quick these days).  I’ll leave it up to you to decide if that’s worth it to you or not.  🙂

So, that’s a very quick overview of Mastered For iTunes, and even though this ended up being a longer post, there’s still a lot of things I didn’t cover.  If you’re curious and want to know more, Bob Katz just released a new book called iTunes Music that goes into more details.  It’s probably a bit overkill for people who aren’t mastering engineers, but it’s only around 100 pages or so and will likely become the bible on the topic like Mr. Katz’s other books.

12 Replies to “Mastered For iTunes”

  1. Jeeez. It never ends w/ Apple does it? I say that as an Apple lover. I use a MacBook, iPod, iPhone and iPad. I love Apple products but not only do they want me to upgrade all my devices every year, spending thousands, now I’m supposed to upgrade my 500-0dd gb. Music collection? I guess it’s a good thing though. Innovation comes with a price. Guess I just need to make more money. Lol.

  2. Yeah, it can get annoying, but I figure once every 10 years isn’t bad. And I get to cull a lot of things ill never listen to again.

  3. No problem, hadn’t seem much talk about this, and I have a feeling it’s going to be something we hear a lot more about soon.

  4. Pingback: “Hooded”
  5. Encode everything into a lossless format and forget about it. How is is are a couple (you want a backup!) of terabyte drives today? 12000 songs (about 1000 CD’s) is about 250GB

    iTunes now lets you downsample to 256KAAC when moving songs back to your iDevice to conserve storage space.

  6. Tarekith, I enjoyed your blog article on MFIT. I transcribe a lot of vintage jazz lps to digital format typically at higher bit and sample rates (24bit/96kHz even higher). Playback is usually through Spin Doctor. With the MFIT format will iTunes now play these versions without downsampling?

Leave a Reply