Clarify that Git objects are created on `git add`

Running `git add` creates a blob object and updates the index. On `git
commit`, a single tree object that points to already existing other
objects is generated and added to object database, as well as adjusting
the refs.

docs/tutorial/file-format.txt

@@ -69,8 +69,9 @@ A blob file looks like this::
 
   blob <content length><NUL><content>
 
-If you change a single line, another blob will be generated by Git at commit
-time. This is how Git can fastly checkout any version in time.
+If you change a single line, another blob will be generated by Git each time you
+successfully run ``git add``. This is how Git can fastly checkout any version in
+time.
 
 On the opposite, several identical files with different filenames generate
 only one blob. That's mostly how renames are so cheap and efficient in Git.