I think folks are thinking of
Thermionic Converters/Generators; right now, we use them to generate power from waste heat generated by radioactive decay in things like space probes, but it's not
directly from the high-energy particles themselves (I think the particles would basically shred any particles/materials we tried to catch them with; one of the reasons plants stick to absorbing low-energy visible light instead of higher-energy UV radiation). Basically, they generate power using the flow of thermal energy from hot to cold, which carries ions across a membrane to drive a current, rather than using heat to boil water to make steam to power a turbine (as in Nuclear Reactors). I'm not sure why we aren't using it more widely; with all the waste heat we produce, we might be able to generate Metric Buttloads of electricity... but I understand no one outside of NASA has really pursued the tech since the 1950s. Which, might I add, is a shame! Insane as it sounds, all reactors from Coal to Nuclear are basically still Windmills (if windmills powered by steam, and generating power using an inverted electric motor). You'd think we'd be past that after a millennium or two, but nooooooo...
So far, the only benefit seems to be a lower construction cost.
A few other benefits of
Magnetized Target Fusion (as I understand it):
1. Doesn't require precision on a nanosecond scale (as with Inertial Confinement)
2. Needs
6 Orders Of Magnitude (1,000,000 times) less pressure to ignite fusion than Magnetic Confinement (Tokamak, et. al).
3. Makes elegant, efficient use of Molten Ferromagnetic Metal to do triple duty; first to generate the magnetic confinement field...
4. ...second, to provide a medium for delivering and distributing energy (compression from the pistons) to the fuel plasma...
5. ...and third, to shield and mitigate the outside area from high-energy particles produced by fusion (No, Neutronicity doesn't magically stop. You can't fuse stuff without high energy particle emissions... it's kind of the whole point).
Sure other implementations of Magnetized Target Fusion have been theorized since the 1960s. We've been working on Tokamak derivatives since that time, and Laser/Inertial Confinement has been in the works since the 70s. None of these are new ideas, and none of them have "worked" yet. It's not the ideas or the theories that we're lacking, but an efficient enough implementation... and that takes time, money, and many iterations, to figure out.
It's elegant, efficient, and sounds awesome on paper... seriously, molten metal, hammers, anvils, and the power of stars. It's basically The Best Image... and honestly, image is an important thing. Science doesn't have to just contend with feasibility and theory, but also with politics and funding. Unfortunately, people not taking Fusion seriously, or seeing it as a pipe dream at best (or magical nonsense at worst) is a problem. Magnetized Target Fusion mirrors the process of pressurizing and igniting fuel in an Internal Combustion Engine in a way that even the everyday non-Physicist can get their head around. Funding further development in fusion will continue to be an issue. Its benefits and efficiency aside, I would not discount the value of Image in overcoming politics, and getting us to Fusion Power.