I’m starting to play around with Mercurial and wanted to see how well it handled FLA files. Mainly I wanted to know how much the repository would balloon as your revised and committed the FLA. So I did a simple test where I made an FLA, committed, added a big image to it, committed, made some other changes, committed, etc. Based on this very quick, simple and unscientific test, Mercurial does a pretty good job storing FLAs efficiently.

Start with empty FLA:
FLA is 57K, .hg is 53K

Added a single image to the FLA (+ the .hgignore file):
FLA is 381K, .hg is 274K

Embedded a font in the FLA:
FLA is 385K, .hg is 283K

Added another image to the FLA:
FLA is 1.1MB, .hg is 1MB

Converted image to MC, added to stage, blurred it:
FLA is 1.1MB, .hg is 1MB (977K)

Changed the blur on the MC:
FLA is 1.1MB, .hg is 1MB (989K)

Made another MC, added to stage, blurred it:
FLA is 1.1MB, .hg is 1.1MB (1010K)

So essentially Mercurial is storing 7 different versions of the same FLA and the repository is only 1.1MB (it’s actually smaller than the FLA it’s been tracking).

Just thought you might like to know.