New
#31
Re. earlier discussion on differentials and incremental backups, I always explain it to people like this.
Supposing you had four files A, B, C & D on day 0. You make a large full image backup with all four file.
On day 1 you decide to add file E, using both methods.
So diff file 1 contains E, and inc 1 contains E.
On day 1 you decide to add file F, using both methods.
Diff file 2 is relative to full image and contains E, F but inc 2 only contains F as it is relative to last inc, so
day 1: diff 1 E - inc 1 E
day 2: diff 2 E,F - inc 2 F
day 3: diff 1 E,F,G - inc 3 G
day 4: diff 1 E,F,G,H - inc 4 H
So diffs grow and are less space efficient. Unless you are storing lots of big files each time, not really an issue.
If say a diff backup file got corrupted in day 2, and you want to get to day 3 status again you can simply use day 3 diff (a little more complicated as files get deleted as well as added but basic principle applies).
However, if inc file 2 gets corrupted, then you lose all subsequent incs as they are daisy chained together.
So diffs offer a slightly higher data integrity. TBH - diffs are easier to manage and free in MRF, and perfectly adequate for home users. Incs are really more for commercial use.