Don't forget the demand side, as we adopt more efficient technologies.
Growth worldwide overall is still far outstripping any shrink in the demand that higher tech. Stuff like the savings from LED house lighting, LED street lights, smart house wi-fi thermostats, and smartphones/tablets replacing desktop PC's etc. will be eaten up many times over as the Third and Second World claws their way up and industrializes and gets utilities. It's going to happen. I mean, you can estimate the energy savings once 99% of the U.S. has LED bulbs, but even if you discount the third world, proliferation in new energy consuming devices is going to eat up the savings.
And barring wars to prevent it, and without major breakthroughs in Solar, the Third World powering up is going to happen via coal power, unless we can commoditize modular meltdown-proof walk-away-safe nuclear power that also mitigates proliferation concerns.
More like ten years off for the last sixty years. Fusion is starting to look like to be one of those pie-in-the-sky technologies that will never be commercially viable, at least in our lifetimes. Maybe after some major technological breakthroughs a hundred years from now fusion power will be a thing.
Fusion is attainable. Because we know it works in stars, and in H-bombs. There's not really some fundamental physics barrier we need to break, it's more of an "engineering problem". Fusion is poised to get commercialized as higher technology in other fields is making it more practical, and enabling us to tackle it in ways we haven't before.
It's kind of like what's happening with SpaceX. Putting aside the love Elon/hate Elon cult of personality and the .gov subsidies, so far SpaceX has managed to do for pennies on the dollar what no one else has been able to, because they've leveraged every new technology advantage we have. Computers, carbon fiber composites, 3D printing and a whole slew of other bleeding edge technologies. I don't think people completely understand the enormity of what SpaceX has accomplished, when one has to look at what their biggest competitors are doing. Things like flying boosters that have a 50-60 year old design lineage, or by buying cold-war surplus rocket engines from Ukraine. And still doing it more expensively.
On top of that, SpaceX is
landing it's goddamn rockets. We've already gotten used to that, but they figured out sustainable reusability in about 3 years, something that took NASA, what... never?
Of course they're standing on the backs of giants to get where they are, and it's not just all "lean" processes designed to minimize bureaucracy and inertia, they have a slew of 21st century technologies at their disposal that previous generations did not,
but that's the point.
Fusion research is starting to get to that same place. Designs like the Wendelstein 7-X stellarator where it's a toriodal confinement ring, like a Tokamak, but it's also twisted or braided, and was a shape that was nigh impossible to make until large scale computer controlled machining and design was possible.
I won't predict commercial fusion in 10 years, I might bet $200 on somebody attaining greater than break-even sustained fusion within 10 years though. And for commercial fusion, I would say definitely less than 100 years.