> And there will probably not
> be much similarity between two versions of an audio file, so I believe
> your repository will grow quite quickly. >
Without knowing too much about WAV files, I would've thought editing 1
minute of a ten minute audio would leave 90% of the file unchanged - so
SVN's binary diffing would be pretty efficient.
On the other hand if you did something as simple as a volume scale
across the whole file, then you'd probably have to store the entire file
Rather than storing the file itself in subversion could you store some
kind "undo log" plus the original file?
ie As you're editing, could you get your audio editor application to
create some kind of small file that says "Cut the audio between X and Y,
Rescale volume by Z%, etc" and store that file plus the original WAV in
subversion (and maybe some intermediate versions if you wanted to save
time on reconstructing files with a lot of changes.
(I don't know of any audio software that does this, I'm just throwing
this out for thought)
Received on Sat Apr 2 04:04:13 2005
This is an archived mail posted to the Subversion Users