AAC, AIFF, Apple LossLess, MP3 or WAV

Mp3, aac or any of the ones you listed will do cd quality-depends on bit rate used and original source.

For +near CD quality+, use AAC at 256 kbps (which is the +iTunes Plus+ setting in iTunes). It is also the current format for iTunes Store purchased music.

For actual CD quality, use Apple Lossless. CD music uses AIFF (another encoding option) which is also lossless, but takes up more space than Apple Lossless. So higher quality taking up more space is not necessarily true. The encoding method also matters.

iTunes rips to Apple Lossless, so it will obviously play it.

The tech specs for the various iPods says Apple Lossless is supported. Some older models may not support it. But you can set it so that iTunes converts the file to a supported format during the sync.

http://support.apple.com/specs/#ipod

WMP? Probably not.

Disadvantage? Much larger file size. You'll be using a lot more storage space, and you iPod will not be able to hold as many songs.

Personally, if I play a songs from a CD and the ripped 256 kbps AAC version on the same audio hardware, I cannot tell the difference.

A couple quick basics just to help clarify some things:

Bitrate is a measure of how many bits of information are being processed per second. For iTunes Store AAC files, that rate is 256 kilobits/sec. (kbps). That rate depends on three factors:

1) Bit depth of the original recording (either 16 or 24)
2) Sample rate of the original recording (from 44.1 kHz to 192 kHz)
3) Compression ratio of the file format (if any)

All CDs are in what's called "Redbook" format. This means the bit depth is 16, and the sampling rate is 44.1 kHz. Uncompressed, that is a bitrate of 4411.2 kbps for a PCM stereo recording. Obviously, to reduce that rate to a mere 256 kbps, you have to throw away a A LOT of the original recorded information. That's what lossy compression is -- you lose information. Now, in theory, AACs and MP3s use complex compression algorithms based on psycho-acoustic principles so that the information that is lost doesn't affect how the listener perceives the sound of the music (again -- in theory). On a relatively low resolution sound system, the differences are not that noticeable. If you have a high res system, however, all compressed formats tend to sound tinny, often harsh, lacking in warmth, have a constricted dynamic range, and much smaller sense of space. My personal opinion as that AACs sound better than MP3s at the same bitrates.

Consider also that many digital recordings were recorded at bitrates higher than Redbook format. Most (if not all) digital is recorded at 24 bits and at least 48 kHz. Many audiophile recordings use sample rates upwards to 192 kHz. For my audiophile downloads at 24 bits 96 kHz, the bit rate is 4608 kbps. When you consider that some of your recordings started life at that bitrate, and then were pinched down to 16 bits, 44.1 kHz, squeezing them down even further to 256 kbps just saps the life out of them.

So, it really depends on your ear and your system. If you wish to be sure you are preserving the full range of recorded sound heard on the CD, go with a WAV, AIFF, or Apple Lossless (which will save space, but without actually removing any information -- just compacting it for storage). Experiment by ripping a CD, encoding one file lossless, and another in a lossy format such as an AAC or MP3. If you hear no difference, well then you can get away with lossy formats. Note, however: if you upgrade your system down the road to the point where you can hear the difference, remember you will have to re-rip your collection. Once you rip a CD and compress it into AAC or MP3 format, you can't go back.

BTW -- for the best resource on the net regarding computer audio, check out this site. They are also very pro Mac over there, so you can get a lot of great advice: http://www.computeraudiophile.com/

Cheers,
Steve

Message was edited by: serdikoff

Hi there Steve,

You mentioned that for your audiophile downloads at 24 bits 96 kHz, the bit rate is 4608 kbps.

I've recently come to conclude that if I try to add WAV files with the same sample size, sample rate and bit rate, respectively, to my iPod nano (installed software is up to date) then iTunes displays an error message saying "cannot be played on this iPod".

Do you happen to be aware if this is a limitation common to all iPods?

Thank you for your time.

AIFF 16bit 44.1KHz stereo is CD quality, period.
Apple Lossless is supposedly CD quality because there is supposedly no data nor audio loss when the lossless file material is compared to the original AIFF (or PCM) file.
The two should be the same in sound quality.

nevadanevada wrote:
I've recently come to conclude that if I try to add WAV files with the same sample size, sample rate and bit rate, respectively, to my iPod nano (installed software is up to date) then iTunes displays an error message saying "cannot be played on this iPod".

Do you happen to be aware if this is a limitation common to all iPods?

Each iPod model has its own limitations on the audio formats it can handle. You can find the specific limitations for all but the WAV format for most models at http://support.apple.com/specs/#ipod. In general, the highest bit rate for the listed formats is a good indicator of WAV format limitations.

Also consider that it is essentially pointless to use higher bit rates & sample sizes than the onboard digital to analog converter of any playback device can handle -- even if the device does internal "downsampling" before conversion, it is highly unlikely that the analog audio output circuitry has been designed with the precision to make any audible difference in the result.

In fact, at different times different iPod models have been considered to have the best "audiophile" sound quality even at lower bit rates than the maximum they can handle because of the specific analog converter & output circuitry they use. Surprisingly, when the original iPod Shuffle was introduced it was considered by many to have the best audiophile sound of any then current iPod.

Similarly, different playback devices vary in how accurately they handle different compression formats, especially the various VBR (variable bit rate) ones. Since this can vary depending on the complexity of musical content, there really is no one format guaranteed to produce the most pristine output, so some experimentation is always a good idea when choosing a format for transfer from your music library to your other playback devices.

Conventional wisdom has been to store the original in the library in its original, uncompressed format (assuming that is available) or an equivalent lossless format & convert that as appropriate for the external playback device. But even that may change in the foreseeable future because of some remarkable advances in methods of recovering what lossy compression formats discard. So far it has only been used with images (pictures & video) but it does what was previously considered impossible: reconstruct missing detail seemingly out of thin air. It is hard to explain how it works but it based on the discovery that if you let a computer alglorhrimically try enough guesses about what the missing details might be, for real world content the simplest guess is invariably the correct one. What "enough" & "simplest" means in this context is the tricky part, involving a highly abstracted mathematical analysis with no everyday language equivalents, but researchers have demonstrated that for a huge number of images, including ones so highly compressed they are almost unusable, the technique can reconstruct an exact or almost exact version of the original.

This literally could change everything we think we know about lossy vs. lossless compression but for now the conventional wisdom offers the best approach.