AAC and MP3 Compared

by Gunnar Van Vliet

Being a Mac user and a music lover, when Apple announced the Apple Music Store for digital song downloads I was very excited. The store uses a new codec called AAC to deliver the songs (for 99¢ each). It’s a competitor to MP3 and since this is the first major attempt at online music downloads for pay, the format is by default the standard for future services. I wanted to know how good AAC was in comparison to MP3, and finally to see if it could come close to standard CD.

I encoded the same track in iTunes using 96, 128, 160 and 192 kbps AAC and MP3, and one AIFF for reference. The track is from the Kansas City Soundtrack - I Surrender Dear. It’s a very well recorded live in the studio jazz piece and it’s a track that I know very well. It features a solo saxophone and trumpet which are clearly localised in the mix and very closely resemble the real instruments.

The usual caveats of testing apply. This is an unscientific test and it’s not double blind but I think that can be good if you’re trying to compare something to a known reference. My stereo consists of Yamaha RX-595 receiver, CDX-490 CD player, and Energy Veritas 2.1 speakers. This system is well balanced and revealing. I’ve compared it to many other systems over the years and it never disappoints.

96 kbps MP3

Harsh ‘digititis’. Extremely rolled off treble. Very phasey. No dynamics. Horrible overall.

Tonal Accuracy - 5/10

Imaging/Sound stage - 4/10

Naturalness - 3/10

Musicality - 3/10

Total - 15/40

96 kbps AAC

Noticeable digital sheen, but overall inoffensive. Very rolled off treble. Poor imaging. Light bass. Slightly phasey.

Tonal Accuracy - 6/10

Imaging/Sound stage - 5/10

Naturalness - 5/10

Musicality - 7/10

Total - 23/40

128 kbps MP3

Flat, compressed sound/dynamics. Rolled treble (quite bad). One dimensional, plodding bass.

Tonal Accuracy - 5/10

Imaging/Sound stage - 5/10

Naturalness - 4/10

Musicality - 4/10

Total - 18/40

128 kbps AAC

Rolled treble, but not too bad. Light bass especially in transients/impact. Compressed dynamics. Surprisingly musical.

Tonal Accuracy - 7/10

Imaging/Sound stage - 6/10

Naturalness - 6/10

Musicality - 8/10

Total - 27/40

160 kbps MP3

Noticeably lighter bass than even lower MP3 bit rates. Smeared/flanged treble.

Tonal Accuracy - 7/10

Imaging/Sound stage - 8/10

Naturalness - 7/10

Musicality - 6/10

Total - 28/40

160 kbps AAC

Decent bass weight. Much better treble definition/air. Somewhat compressed dynamics. Good, but still a little lifeless.

Tonal Accuracy - 8/10

Imaging/Sound stage - 8/10

Naturalness - 8/10

Musicality - 7/10

Total - 31/40

192 kbps MP3

Mid-range somewhat forward. Good imaging.

Tonal Accuracy - 7/10

Imaging/Sound stage - 9/10

Naturalness - 9/10

Musicality - 8/10

Total - 33/40

192 kbps AAC

No sparkle. Light bass, although with good detail.

Tonal Accuracy - 7/10

Imaging/Sound stage - 7/10

Naturalness - 7/10

Musicality - 7/10

Total - 28/40

AIFF (in comparison)

Beautiful sparkle to piano keys. Generally filled with much more life and atmosphere on a tactile level. Far more musically involving. This is the reference piece so it naturally gets a perfect 40/40 score.


On the whole, there weren’t any surprises. My observations echo what most people have said about AAC vs. MP3. AAC is higher quality at the same bit rate, so you can use a smaller file to achieve the same quality as MP3 which is a good thing for portable and computer users. Ultimately, both formats still sound pretty bad in their practical ranges compared to CD. I didn’t test 256 or 320 kbps because it’s impractical for most users to use these encodings. The Apple Music Store for example uses 128 kbps, and if you have room for 320 kbps and you care about sound that much you’ll probably use AIFF or just play the CDs themselves.

You’ll notice a few anomalies in my findings, such as higher bit rates from the same format getting a lower score, and one case where AAC did worse than MP3. These I attribute first to the interaction between the piece of music chosen in relation to the codecs - sometimes less data sounds better, or more distortion sounds better if it gets the right mix of psychoacoustics. An analogy might be to vinyl or tubes, or even compressed FM radio - getting more of the good part of the music, ie. the fundamentals and less of distracting ambiance/texture can actually sound better, or just more enjoyable sometimes. Secondly, as fidelity increases, flaws, or what’s missing can become more apparent. An analogy might be to HDTV. When I first saw HDTV I found it blurry (although much “clearer” than regular TV) because I had jumped exponentially in expectation. I wasn’t comparing HDTV to normal TV, my brain had jumped standards to compare it to real life!

In any case, these are just some of my thoughts. I know I’ll be sticking with CDs and LPs for a long time.


I’ve received tons of e-mail thanking me for the test, and many people have suggested that I use OGG or MP3 with LAME/VBR because they’re better than the iTunes standard encoder. I’ve read many comparisons using different codecs and it’s probably true, some are better than MP3 or AAC, but that’s not the purpose of my test.

The vast majority of Mac users (and perhaps Windows users soon) use iTunes. Moreover, that’s what the Apple Music Store is based on so it makes sense to me to look at what is the defining standard for online music downloads. This is important to keep the bar high for these services in the future. People will always come up with ways to improve their encoded files and these are very legitimate but the majority of people will never even open the preferences.

When the iTunes Store becomes available in Canada, I will conduct a Part 2 test using one of Apple’s files to benchmark against. I’ve gotten so many requests for this, that if someone can contact me regarding obtaining a reference file that I already have on CD, I can go ahead now. I will also do an OGG, and MP3 LAME/VBR comparison with bit rates all the way up to 320 kbps.