Say I copy my CD from 1996 to my computer. Then a year later I forget and copy it again. Now I have two folders with the same tracks. Same “exact” software, same settings, maybe different versions of ffmpeg, maybe different CD/DVD drives, maybe different OS versions. Filesizes are close but not identical. All files together are 297 MB versus 298 MB. On playback they sound the same. Spectrum graphs look the same, VLC’s Media Info / Codec dialog is identical. What other than A/B listening is useful to determine which copy to keep, according to my values: FLAC or other lossless, 24/48/96 K …
When copying (ripping) a CD, you typically wouldn’t use Audacity. That’s a job for a tool like Exact Audio Copy or CUERipper. Both of those copy the exact digital bits off the disc, where for Audacity (as far as I know) you’d have to use an analog loopback which isn’t the right tool for the job.
When copying the digital bits, in theory both versions should be identical, as long as you’re comparing an uncompressed (i.e. WAV) file. This is just a theory, however, because the original CD might have gotten more scratched since the last time you copied it, or the reflective coating deteriorated, so in those cases the digital rip might not be identical.
As soon as you add lossLESS compression like FLAC, they might be different because the FLAC tool might be a different version which has bug fixes or improvements, or the metadata might be different.
If you did try to make a copy with Audacity using an analog loopback, that’s almost guaranteed to be different every time.
1 Like
The situation that most closely matches your situation of different audio file sizes is that they have different metadata tags embedded within them.