0

Okay so my understanding is that each time I render/export audio as anything it compresses itself and u lose quality. Even exporting it as itself with no changes once more, loses quality. That its similar to video codecs but less noticeable loss. That choosing a lossless format contains the most data and therefore will lose the least amount of quality. So assuming these are correct statements...and I "have" to render audio out multiple times to get it into the form needed (like when doing video editing projects for example), what formatting types are acceptable/best practice to avoid quality loss?

We'll assume we aren't mixing audio, just re-rendering the same one over and over with slightly different start points lets say (so like my 2nd render the audio starts playing at 2s, my third render it starts at 5second mark, etc).

  • If you're merely changing the edit point, aren't you just always starting from the same original footage & audio track? How are you getting generational loss? – Tetsujin Nov 23 '18 at 7:51
1

You can easily test this and prove to yourself that this is not the case.

Create a single track with a sine wave and export this to a mono WAV file.

Import the WAV file you have just exported to a new mono track.

Invert the phase of this new track and play both tracks together so that they mix.

You will notice that there is zero output, therefore there has been no degradation during the export process as there has been no change to the track during export.

EDIT:

Three broad categories of codec:

  1. Full-Bitrate - where all audio data is retained, no bitrate-reduction is attempted and no psychoacoustic data-rate reduction is attempted. Examples: PCM
  2. Non-Lossy - Codecs where data-rate reduction is performed, but there is no change to the audio data after a full code-decode operation. FLAC, WAVPACK, APE, ALC
  3. Lossy - Codecs where data-rate reduction is performed using psychoacoustic masking and other techniques to remove audio data that can be safely masked. Significant data-rate reduction can be achieved with little loss of quality. Data-rates can either be constant or variable. Examples: MP3, AAC, OGG, OPUS
  • Oh that's cool. I guess I'll use that as a baseline to see if loss has occurred at any point then. Is this exclusive to lossless formats though (.wav and flac)? And is it independent of bit rate? Or do lossless formats have fixed bit rates (my programs present differing information on that)? Like i tried it with mp3 and rendering mp3 as mp3 you hear the difference vs the inverted original...which confirms what I read about re-rendering lossey formats. My audio programs all differ in the settings and information so its hard to figure out what one is giving placebo info and which isn't. – kite Nov 24 '18 at 2:54
  • lossless formats do not necessarily have fixed bitrates. You can perform data rate reduction alongside a codec being lossless. Codecs such as PCM are constant and full bitrate codecs. They perform no data rate reduction. FLAC, APE, ALC, WAVPACK are all lossless codecs that perform some level of data-rate reduction. My answer refers to DAW's that export as WAV. – Mark Nov 24 '18 at 8:04
  • @kite - further to my edits above, any exporting to a full-bitrate or non-lossy codec will be fine. Lossy codecs will cause degradation after multiple coding cycles as you have described. – Mark Nov 24 '18 at 8:31

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.