Re: [PATCH 3/5] pack-objects: add --sparse option

+--sparse::
+       Use the "sparse" algorithm to determine which objects to include in
+       the pack. This can have significant performance benefits when computing
+       a pack to send a small change. However, it is possible that extra
+       objects are added to the pack-file if the included commits contain
+       certain types of direct renames.

As a user, where do I find a discussion of these walking algorithms?
(i.e. how can I tell if I can really expect to gain performance benefits,
what are the tradeoffs? "If it were strictly better than the original,
it would be on by default" might be a thought of a user)