According to standard proofs found in any analytic thermodynamics book, at constant T:
$$ dF > dw $$
ie, the change in the Helmholtz free energy gives the maximum amount of energy which is free for a system (at constant T) to convert to work.
However, this is frequently followed by the statement comparing it to Gibb's free energy and stating that "dF also includes the unextractable (and hence useless) pV work, in contrary to dG at constant T and P".
How comes work by volume change is considered useless? Isn't that exactly what we normally use to drive pistons and engines in most standard thermodynamics problems?