Last week, General Fusion abandoned at least 25% of its staff at least a few days after its critical milestone in its latest Fusion demonstration equipment.

CEO Greg Twinney posted an open letter on the company’s website on Monday, which has less universal fusion, although its new LM26 device is capable of compressing plasma (a necessary condition for fusion). "

The 23-year-old company is still described as a startup and is the leading contestant in Canada's business convergence power competition. According to PitchBook, it has raised $440 million, including 22.66 million rounds closed in July. Supporters include Jeff Bezos, Temasek and BDC Capital. However, the money is not enough to make the company show that its unique convergence method is feasible.

Fusion General's dilemma highlights the challenges facing the convergence industry.

To date, only one device has hit the so-called scientific break-even, which is historically important but not very important milestone. To hit the commercial breakeven, reactors need to be dozens of times more energy than proven so far.

The road to these milestones has proven to be costly. Although the Tally of Federal Fusion seems impressive, it is in the middle. Commonwealth Fusion Systems has raised more than $2 billion, Helion has earned more than $1 billion in revenue, and Upstart Pacific Fusion has promised $900 million in its Series A alone.

Part of the Fusion challenge is that it takes a different approach than many of its competitors.

Most Fusion startups follow one of two paths: magnetic limit or inertial limit. The former uses a magnetic field to control the plasma and squeeze it to meet the conditions for nuclear fusion. The latter method typically uses laser compression of fuel particles.

On the other hand, ordinary fusion has been trying to compress the fusion fuel using steam-driven pistons. The U.S. Navy tried something similar in the 1970s, but it was useless, but ordinary fusion believed that modern computers could solve some of the timing problems that plagued early attempts. The situation has not been shown, but it says that if done, LM26 should be able to achieve scientific breakeven.

Now, if a company wants to prove that its approach is a viable competitor, the company will have to raise more money quickly.