Wierd! Looks like the angle between the connecting rods and the piston travel is REALLY large on the power stroke. Doesn't seem like it would be very efficient.
That's what I thought initially, but does it say which direction it spins? I don't know if there'd be a net benefit either way though.
And actually, there's no more travel on any part of the stroke than a normal engine. If you pretend that one crank isn't there, the travel is the same except for the exaggerated unevenness of lateral forces on the piston. Add the second crank into the equation, and you've taken out lateral forces all together, eliminating one source of friction based inefficiency a conventional engine has, and adding the friction of another set of bearings on another crank.
So thinking through all the sources of friction and rotational mass involved leads me to conclude that I'd need a degree in engineering (I wish!) and some math (yeah right) to figure out how well that design would work.