Typically you want the vibration damper or absorber at the OPPOSITE
end of the crank from where you're taking the power.  That's the way
it is on automobiles where you've got the flywheel/transmission at
one end of the crank and a viscous damper at the other.  The latter is
often combined with the crank pully (for driving the accessories).

The crank "winds up" in use - it twists.  The greatest phase difference
("twist") in the crank will be measured at opposite ends of the crank.  Since
you want to minimize or eliminate this twisting of the crank you put
something at the opposite end of where you're taking your power to try and
make the other end "keep up" - i.e. keep in phase - with the power end.

There are various devices used to take care of crank twisting (torsional
vibration) in service.  By far the most common is the aforementioned
automotive damper.  This simply works by applying a retarding force to
the crank which tends to lower the magnitude of its acceleration and
decceleration.

Note that you cannot do this with a rigidly coupled device, such as a flywheel.
You must have an elastic coupling between the crankshaft and the damping
weight.  

The automotive damper takes the vibratory energy of the crankshaft and
dissipates it as heat.  It is a net energy DRAIN on the engine.

Aircraft engines use a sarazin or chilton type pendulous damper.  This
device stores the acceleration of the crankshaft as potential
kinetic energy and forwards it back into the crankshaft when the
crank starts to deccelerate.  It accomplishes the same thing that
the automotive damper does but it uses a different principle.  It also
does not remove energy from the engine - hence it does not heat up.  Instead
of dissipating torsional vibration as heat, it stores it and returns it
to the crank.

Now torsional vibration is probably the MOST vexing problem associated
with reciprocating engines.  That's all kinds and all applications (ships,
cars, planes, etc.)  We've gone over what it takes to begin keeping it
in check inside the engine proper.  But there are still lots of problems
with it escaping out of the engine and into your gearbox, prop shaft,
prop, etc.  You may have all the damaging vibration modes for the
CRANK damped within the engine but there can still be vibration orders
which escape the engine and excite things that you've bolted to it.

Jess was no doubt referring to the latter.  If you can avoid it, you really
don't want gas pressure pulses to get into your PSRU and/or prop.  One
way to do this is to use something like a hydraulic torque converter between
the crank and the PSRU.  Again, this does nothing for crank vibration per
se, it just keeps it within the engine.  You STILL need some kind of
absorber on the crank itself.