"I am not sure how someone can see something the size of a coffee can that can out at least 14 volts at over 100+ amps while still being air cooled inefficient."
The efficiency he was referring too is not the relationship of space to power output, but rather the relationship of input power to output power, which is in fact very poor, probably because the auto companies are more concerned with low cost than high efficiency, and they have an excess of power available. I have a setup with an older GM alternator driven by a 3 HP B&S horizontal shaft eng with a pulley ratio of about 2 to 1 (on the vehicle I took the alt off of the ratio was about 3:1 but my 3 hp lacks enough torque to drive it like that) So at cruising speed the alt was turning about 7500 rpm and I figured their eff curve was based around that RPM. Anyway 3 Hp petrol engine will only output about 20 amps and then it loads the engine down and kills it. I have used two different engines one an OHV engine, and one a flat head, and both are about the same, but as expected the OHV would do a little better, but not much. I am an electrical engineer and I fully understand the Physics behind 746 Watts = 1 HP, but based on empirical data it appears that Alternators are only about 10% eff, however I suspect also that the engine manufacturers are also fibbing a bit, and you also have losses in the belt pulley system, so I would say maybe 25% overall eff on an alternator.