An IEC (Farnsworth Fusor) would be a good place to start for the neutrons but you will need fuel. The poster seems keen on doing everything himself and breeding fuel should be no exception. I recommend starting with Deuterium and working your way up to breeding tritium. You can buy deuterium easily enough, but everyone knows that path is for chumps. I would say you should get a good centrifuge cascade going and start separating heavy water out from the normal stuff that comes right out of the tap. With a little electrolysis setup you can pull the deuterium right out of the heavy water! Put the Deuterium into your IEC and snap, its a neutron source. Pro tip: wrap a few layers a beryllium around your IEC for neutron multiplication, but don't eat it kids (its toxic). If you want to step up to D-T fusion you can add a breeding blanket of Lithium into the mix. Separated Tritium from the blanket can go right back into the IEC for extra neutrons.

Safety note: if at any time during this exercise you feel like tiny knives are murdering your sperm don't worry, only the weak and unworthy sperm will be purged!

The problem is not as easily solved as you make it out to be for a few reasons. The first being demand and the second being supply. The article doesn't really go into much detail but the real demand issue is the rising use by of He3 by the US gov in portal monitors. He3 tubes are by far the best devices available for neutron detection. Since 9/11 the US gov demand for He3 neutron tubes exploded and pretty much ate the entire stockpile. This has caused major headaches for everyone who uses He3 like the medical field and basic science research.

On the supply side He3 is created when tritium decays on a 12 year half-life. The largest supply of this for many years was the US nuclear weapons program. Production now, however, is nothing like it used to be. Without the tritium production we don't have the He3. Even if we did we might not meet the kind of demand we have for He3 now. In order to make 1kg of He3 you need to let 2kg of tritium decay for 12 years. Or you need to let much larger quantities of tritium decay for shorter periods of time. Either way you need a lot more tritium than we have.

Additionally getting He3 from heavy water reactors is probably not an option. The best way (the way the US gov does it anyway) to make tritium presently is by putting lithium rods into a reactor and then removing the tritium from the rods (its a fission product from lithium). While tritium is produced in heavy water reactors by neutron capture, the cross sections are very low. This mean you would need to separate the heavy water out from the tritium rich water (centerfuges) and then remove the tritium form the water molecules with electrolysis and then again separate tritium from deuterium. This ignores the fact that All the commercial reactors in the US are light water (normal H20) and countries that use heavy water (Canada) may not be interested in stockpiling tritium.

Production difficulties aside tritium is just plain expensive. tfa cites the He3 price at $5000 a liter with a goal of more like $1500/L. This puts the price roughly $37500 a gram. Tritium is presently $25000+ / g and that is a subsidized price. Its estimated that actual production cost is upwards of $75000 / g

Given all this, if we had a cheap easy solution laying around we would have done it by now.

I think perhaps we are looking at this wrong. I can think of reasons why it might or might not work, but it is perhaps more productive to think about what kind of games I would want to see produced.
I would probably pay more for some games than I would for others. It would also help if they asked for the money after securing a development house I trusted.

Games I would want to see ( a lot of these could be done with existing game engines)

    A new x-wing/tie fighter game (it has been way to long) (maybe with rpg elements)

    possibly a star wars rebellion sequel (grand strategy)

    Something new from the Jedi Knight series (force unleashed doesn't count without a pc version)

    A new x-com game

    I am sure there would be plenty more if I dug through my old favorites

Clearly ip becomes a major issue. Much of what people might want enough to pay for is in sequels for games (or ip) that they already loved.

Obligatory post

Inertial confinement fusion does not rely on having a stable plasma for any extended period of time as magnetic confinement does. Instead, think of it as a series of small bombs. Each is fired into the center of the chamber and ignited with the laser system. In a commercial plant this would have occur 5-8 times a second. Meaning you have what is essentially machine gun speed firing of DT pellets into the center of the chamber with equavalent speed lasers. Thus one of the large problems remaining in ICF fusion is the development of the laser components that can fire in this way for extended periods of time. Additionally, first wall materials are needed that can handle the neutron and ion flux that is generated in extended operation. The major US project that was actually addressing the laser and material tech side was HAPL, which got zeroed out on the FY 2009 budget.