Fonestrap

[cacophony]

aac is compressed, so you've already got less data than the original file. if you then convert to mp3 you lose a little bit more of the information. it's basically cascading lossy compression. the ideal situation is to do as described above, where you leave the compressed file unaltered and just essentially pass through to a different file extension.

of course not everyone is going to be able to figure the process out so just keep in mind that the cascading is worse if you compress twice with the same algorithm, and is more likely to be audible. if you absolutely must convert, choose a different algorithm to try to hide as much of the audible loss as possible.