Short version:
FLAC is a lossless format. Whether you export as FLAC or as WAV, the sound will be identical.
Do you ever use ZIP files?
I’m starting to have some space issues
I’m more concerned about that. Audio (and Video) production doesn’t “do” space issues.
“For Some Reason Audacity crashed and I can’t get the show I’ve spent months editing to wake up.”
Under many conditions Audacity makes a copy of the whole show as UNDO. Can you support multiple copies of your multi-hour show in Audacity’s super high quality format? What else is running on your machine that may need space? Some programs such as Skype and Even Windows don’t ask. They just take.
You need to be crystal clear about the risks.
There is one thing worse than the above.
“… and I have no backups.”
Koz
Sure. There could be disk errors, faulty RAM, hardware problems… What there won’t be are errors resulting from the encoding / decoding algorithms. ZIP files have been around for 30 years, and used with all manner of data, from audio, video and images, to financial data, scientific data, all kinds of data. Lossless data compression, is in effect, two mathematical equations. One equation does the encoding, and the other does the decoding. The decoding algorithm is the inverse of the encoding algorithm, like “multiplication” is the inverse of “division”.
If A x B = C
Then C / B = A
This is always the case. There’s no way that C divided by B can be anything other than A.
Similarly:
PCM data → FLAC encoded => FLAC file
and
FLAC file → FLAC decoded => PCM data.
Just as:
PCM data → WAV encoded => WAV file
and
WAV file → WAV decoded => PCM data
FLAC files just use a different codec than WAV, and the FLAC encoded data is squashed together whereas the WAV encoded data is evenly spaced out. The “data” is the same.
An advantage of FLAC is better metadata support. Metadata is not well-standardized for WAV. To me, that’s more important than the space-savings.
FLAC also includes a checksum so the integrity of the file can be validated. (I don’t actually know how to check that, or if it’s checked automatically when you play/decode the file.)
On the other hand, one bad byte in a WAV file will corrupt only one audio sample and a bad byte in a compressed file can corrupt multiple samples. And, since a WAV file is simply a header followed by a series of samples, if the header is damaged the audio can be completely recovered, or if any part of the audio data is corrupted the remaining good-audio can be recovered (assuming the file can be read).
On Windows you can test the integrity with the “FLAC Frontend” app: http://flacfrontend.sourceforge.net/
On Mac and Linux you can install the command-line file encoder/decoder and test with the -t option:
flac -t "name-of-file.flac"
Except for B=0
I was of course referring to valid arithmetic expressions.