Fascinating how much variance there is in the power we can derive from various raw materials. Last week I noted that we must burn 1000 tons of coal to supply the energy needs of an average American for a year, and that the same amount of power could be generated in a fusion reactor fueled only by the deuterium in half a gallon of water and the lithium in a watch battery. Looking a little further we see that’s roughly the same amount of power we get burning 4000 barrels (168k gallons) of crude oil. Fission reactors, which can work on various isotopes, but currently which are mostly designed to run on uranium, can generate the same amount of power with just 1 pound of uranium-235. But since natural uranium is only 0.7% U-235 we would need 700 pounds (320kg) of raw uranium to produce that power.
(If we ever managed to construct an antimatter reactor, we could get the same power by reacting just 10mg of matter with 10mg of antimatter.)
The variance in productivity given a unit of energy is also interesting. According to this page the average human over time consumes/emits the same amount of energy as a 100W light bulb left on all the time. Consider that the latter has the ability to illuminate just a single room, while the former can both reproduce itself and create/control many orders of magnitude more energy. (Or consider that current CPUs can compute as many as 10 GigaFLOPS with the same power!)