日本語 
Fujitsu Limited and The University of Osaka are trying to solve a problem that has been hanging over quantum computing for years. It promises a lot, but in practice, it is still not usable where it actually matters.
Their latest work focuses on early fault-tolerant quantum computing. Not the ideal future version with millions of qubits, but the messy, limited systems we will have first. That is where most approaches struggle. Either the calculations take too long or the resources required are unrealistic.
What they have done is combine an upgraded version of their STAR architecture with a new way to optimize molecular models. STAR, in simple terms, is their way of making quantum computations more efficient, especially when it comes to phase rotations. The new version pushes that efficiency further. On top of that, the molecular optimization technique reduces the load even more.
Also Read: Mitsubishi Heavy Industries Expands Eco-Friendly Heat Pump Portfolio to Capture Europe’s Green Energy Demand
The result is not small. Calculations needed for chemical material design, like catalyst molecules, can now be done in a realistic timeframe on early-stage quantum systems. Earlier, this was not just difficult. It was practically impossible. Even older versions of their own architecture would take an absurd amount of time.
This matters because energy calculations sit at the core of things like drug discovery, ammonia production, and carbon recycling. If you cannot compute accurately and fast, progress slows down.
Step back and look at the direction. The industry is shifting from theoretical breakthroughs to usable ones. Not perfect systems, but good enough to start solving real problems. Fujitsu and Osaka University are clearly betting that the first wave of quantum value will come from working within constraints, not waiting for perfect machines.
English