The idea of scaling has long shaped the technology industry. Moore’s Law gave the world a roadmap for faster, smaller, and cheaper electronics, and for decades, the industry delivered. But quantum computing presents a different kind of challenge, one where raw numbers matter far less than practical performance. The hard part is not just adding more qubits. It is making them work together reliably and repeatedly. One participant in the SPIE panel, Erik Hosler, known for his work on advanced patterning challenges, pointed to the critical difference between building systems that look powerful and building systems that are powerful.
Quantum researchers are now confronting a wall. They are finding that going from a few dozen high-quality qubits to hundreds, and then to thousands or millions, is not a straight line. It’s a cliff. This scaling wall defines the limits of what current architectures, fabrication methods, and error correction schemes can achieve. To make quantum computing viable, the entire system has to improve, not just the count.
Why Qubit Quantity Isn’t Enough
In quantum computing, qubit count is often used as a headline metric. A 127-qubit system might sound better than a 65-qubit one. But in practice, the number is only meaningful if those qubits are stable, controllable, and error-resistant. Most experimental machines today suffer from short coherence times, gate infidelities, and noise from surrounding systems. These problems compound as the system grows.
The result is that more qubits often lead to more problems. If they cannot maintain their state long enough or if they produce error-prone results, the system becomes difficult to use at all. And that is before accounting for the overhead required by quantum error correction, which demands additional physical qubits just to maintain one logical qubit.
What matters is not just having many qubits. What matters is whether they can work together to perform deep, sequential operations reliably. That is the key to unlocking problems like cryptographic factoring, complex simulations, or high-dimensional optimization. Usability, not sheer quantity, determines whether a machine can shift from theory into practice.
The Scale That Actually Matters
This distinction was made explicit during the SPIE panel session. “We need hundreds to thousands of usable qubits with the capability to do billions of sequential operations to really do useful work,” Erik Hosler emphasizes. It reframes what counts as meaningful progress in quantum computing. It is not enough to build machines with theoretical capacity. They must be able to execute long operations, chain logical instructions, and maintain coherence throughout. Depth and durability are now just as important as raw scale.
A small system with a few exceptionally reliable logical qubits may outperform a larger machine that struggles with noise and control. And until researchers can reach operational benchmarks that support real-world workloads, quantum will remain largely experimental. That is why usable qubits, those that can sustain billions of uninterrupted operations, have become the new goalpost.
It introduces a fundamental rethinking. Developers must shift attention from short-lived demonstrations to sustained operation. The hardware, software, and control logic must all support workloads measured not in seconds, but in sequences numbering in the billions. It is not an academic ideal. It is the baseline for running simulations that can model complex molecules, navigate non-linear supply networks, or optimize quantum-chemical reactions. None of those tasks can be meaningfully performed on systems that falter before completing a sequence of steps.
The Fabrication Challenge
Several speakers at SPIE addressed the growing difficulty of precision in chipmaking. That challenge becomes even more pronounced in quantum systems. Arrays of qubits must be fabricated with nanometer-scale accuracy, using materials that are extremely sensitive to environmental interference. Any variation can disrupt coherence or introduce phase errors.
It creates a new kind of manufacturing bottleneck. Quantum devices are no longer lab experiments soldered together by hand. They are semiconductor-class systems that must scale through reproducible fabrication techniques. Patterning accuracy, materials uniformity, and interface stability all play critical roles in determining whether a qubit is functional at scale.
To complicate matters, many quantum platforms must operate at cryogenic temperatures. That constraint affects everything from packaging to wiring to testing. Engineers are now tasked with creating devices that maintain fidelity under conditions that push conventional processes to their limits.
Parallels exist to the advanced lithography techniques discussed at the symposium, particularly in resist design, defect management, and multilayer integration. These fabrication capabilities will likely play a pivotal role in making large-scale quantum processors possible.
As the industry shifts from small prototypes to scalable architectures, fabrication must become part of the performance equation. Usability at scale depends not just on qubit design, but on repeatable, industrial-grade manufacturing.
Getting Beyond the Numbers
There is growing recognition that performance metrics must be developed. Qubit count alone no longer tells a meaningful story. What matters is the number of usable qubits, how long they can retain their state, and how many sequential gates can be applied before the system breaks down.
That realization is reshaping development strategies. Some groups are prioritizing modular architectures that can cluster high-quality qubits into interconnected blocks. Others are working on topological qubits designed for built-in error protection. Still others are exploring hybrid systems that combine classical control with quantum processing layers.
What unites these efforts is the goal of pushing past the current wall. The problem isn’t that more qubits are unavailable. It’s that scaling without stability produces diminishing returns. Moving from experimental setups to machines that can run real algorithms for real customers means changing how progress is defined.
This transformation may also shift funding models. Investors and research backers will demand systems that deliver repeatable, extended performance. Academic demonstrations will no longer be enough. Quantum must mature into a production-level model if it’s going to justify its promise.
Quantum’s Future Is About Durability
If quantum computing is to transition from scientific novelty to industrial asset, it must demonstrate more than architectural ambition. It must offer practical capability. That means reaching the level of performance described, systems that can support large numbers of useful qubits and sustain deep operations without failure.
The companies and research teams that get there first may not have the biggest machines. They will have the most durable ones, machines that can manage messy, multi-step problems and repeat results reliably, on demand.
Industry benchmarks are already beginning to reflect this shift. Metrics such as circuit depth, gate fidelity, and error rate are gaining weight in system evaluation. The trend is clear: Scaling is no longer just vertical. It’s about how far you can go horizontally, how long the system can run before it breaks down.
Use cases will follow that logic. In pharmaceuticals, quantum may shorten the time to drug discovery. In logistics, it may make supply chain planning more adaptive. But only if the systems running these models can survive the complexity of the problem itself. That is the gap between theory and delivery. That is the real wall that quantum must cross.
The future will not be decided by who reaches one million qubits first. It will be decided by who reaches one thousand usable qubits that can work without breaking under pressure.
