Basics: Ripping

[Dan Gravell]

I wouldn't use AIFF - the tagging support exists but is patchy in much software/hardware.

 

Go for FLAC or Apple Lossless (ALAC, if you are tied to the Apple ecosystem).

[Dan Gravell]

Don't see the advantage of recommending a losslessly compressed format (ALAC, FLAC) over an uncompressed, lossless format (AIFF).

Hard drives are so cheap nowadays that storage space should not be the issue. AIFF is supported by all platforms and virtually all music player software, while FLAC is not supported by iTunes or any Apple device. ALAC is currently not supported by HDPlayer, one of the best, if not the best, music player program available.

 

Because there's no cost other than a little CPU in the [de]compression.

 

I agree storage space isn't really an issue, particularly for "archiving" purposes. But my point wasn't so much to laud compression as to question the use of AIFF with, as I said, 'less good' support for metadata storage.

 

It comes down to what makes sense for you, but as general advice I wouldn't use AIFF.

[Dan Gravell]

I don't get the argument that lossless compression makes a CPU work harder converting during streaming and thus a slight loss of SQ, when folks are now over sampling to 8X DSD. Now that's making the CPU work harder.

More to the point, I don't think it's relevant to discussing archiving.

[Dan Gravell]

Away from the subject of file format, my other recommendations would be...

 

Store as little metadata in the file path as possible. Just identification tags - artist/album/tracknum-trackname with the media number (disc number) in there also as required. The reason is that changing file paths is the most invasive change to a music library; it can screw up software that use file paths as identifiers, makes synchronisation harder, etc etc. The less data in the file path, the less likely you'll have to change it.

 

Ideally enforce formatting rules of tags, e.g. date fields should use the same date format, tracknumbers should all be padded or not and so on.

 

Err on the side of storing fewer classification/categorisation tags, and more identification tags. The reason is that the former are more likely to change according to your whims, while the latter are always useful for finding data later. But your CD ripper may not allow you to control that.

 

Backups are so important!