It's probably also worth noting that it's realtively easier to achieve high VE and also hang on close to peak VE longer in a relatively low(er) rpm engine, really regardless of displacement.

The reason is that with a 4 cycle engine the frictional losses(rings/bearings/valvetrain spring pressures and lobe friction) and the drag from the rotational masses all increase at the square of the RPM....so it's like applying an increasingly linear load to the crank as RPM goes up. Kind of like a tractor pull sled, the further up you go, the more drag is applied that the heads and charge have to overcome. The higher you want to rev it, the stiffer springs and stronger/larger (generally) components you need to get there. The only way F1 motors rev as high as they do (in excess of 22,000rpm) is because they don't have conventional valvesprings for the crank to have to overcome. Imagine how much power you would have at the wheels if you camshaft and valvetrain were driven by an electric servomotor instead of a crank and chain driven camshaft. That's the way things have been moving and it's remarkably simple to do today.

And also....generally, after peak VE RPM occurs (for a given motor) the RATE of crank acceleration begins to slow in terms of RPM (or probably more accurate to say Revs/Second per second. That's why hanging on to (as close to) peak VE allows the power curve to continue to rise. And with a flat torque curve the rate of rise is essentially linear (rises proportional to RPM increase )

The optimum port size/shape and valve timing is all relative to the size of the engine AND (more importantly) the range of RPM you want to apply load to that engine. Obviously compression and induction/exhaust optimization all come into play as well.