That video makes it sound like we're far away, because we need tokamak output to be about a hundred times better than we've achieved. What it ignores is that t

That video makes it sound like we're far away, because we need tokamak output to be about a hundred times better than we've achieved. What it ignores is that tokamak output is highly nonlinear. It scales with the square of plasma volume, and the fourth power of magnetic field strength. Double the field, 16X the output. These scaling laws are very well established at this point. The plasma volume scaling is why ITER is so big. But after ITER was designed, people invented REBCO superconducting tape, which can support much stronger magnetic fields and is commercially available. Using those fields, CFS is building their SPARC reactor, which will get ITER-level output (Qplasma=10) from a reactor half the size of JET, which was built in four years. (Three years for the building, one year for the reactor inside.) If that goes well, the next step would be build their ARC reactor, which will be the same size as JET and get commercial-level output. (Some of the alternate designs also have great scaling in theory, but we don't understand their plasma physics as well so it's harder to predict how they'll turn out.)