The world currently consumes 450Q(BTU) of manmade energy per year. This equals energy of 5500, one-gig nukes, or 16,500,000, one-meg wind turbines (at 8-hours per day, 3 gigs of rated wind, equal one-gig of nuclear). At $4B per nuke, or $2M per turbine, 450Q would cost $25T or $35T respectively. Better yet, for $60T capital cost, plants with 900Q yearly energy capacity could be built. Costs would rise thru the century, but nowhere near oil inflation; fuel will always be cheap (one pound nuke fuel equals 500,000 pounds coal, and wind is free).
These capital costs may sound high, but they are bargains compared to current $1.8T per year for $60 oil (85MBPD, equal to 30BBPY). (This equals $18T in just a decade.) As oil disappears this century, long before, most likely, oil will be $100, $200, $300 and more per barrel (true world price without speculation). Yearly costs will easily be $3T, $6T and $9T, respectively for only 40% of 450Q. Per decade, if indeed the world can afford it, oil will cost $30T, $60T or $90T.
The $60T capital cost (as escalated) would be easy for the world during this century, and would give 900Q of clean energy which is FIVE times as much as current 170Q of oil. Even as capital costs rise through the century, these plants will always be massive bargains; and incidentally, save world civilization from climate change.
In a later post, possible bottlenecks and construction difficulties will be discussed.