Calculating the "wasted space" for a given cluster size:


The rule of thumb is to take the number of files times the
cluster size times one-half.

The cluster size does make a difference!

The idea is that - on average - half of the files will use
less than one-half of a cluster (for the very last cluster)
and half of the files will use more than one-half, i.e. the
sizes will vary randomly but average out to about one-half
of a cluster.

As any rule goes, there may be exceptions.  For example, you
may have a large number of files that use only a small part of
a cluster (for the last one).  In this case you would waste
considerably more space.

Dean Kukral