German startup unveils first room-temperature diamond quantum computer past 10 qubits

[StringTheory97]

A German startup called Saxon Q has debuted what it says is the worlds first diamond based quantum computing system to exceed 10 qubits, and the genuinely wild part is this thing runs at room temperature and just plugs into a normal power outlet

The machine uses whats called a nitrogen vacancy architecture, meaning it exploits naturally occurring or lab grown defects in synthetic diamonds where a nitrogen atom sits where a carbon atom should be, creating a vacancy whose trapped electron can be manipulated with lasers and microwave pulses into quantum states that ordinary binary bits cant achieve

Marius Grundmann, a physics professor at Leipzig University and Saxon Q co-founder, said the breakthrough that let them push past the 10 qubit barrier was a materials discovery, specifically co-implanting sulfur alongside the nitrogen vacancy, which lifts the chemical potential enough to make the vacancy negatively charged and dramatically increases the yield of usable qubits

The current hardware ships in rack mounted systems up to 128 qubits with 512 qubit configurations coming next year, and Saxon Qs roadmap targets 10,000 qubits and beyond after 2030, and Grundmann reported single qubit fidelity of 99.98 percent as of late July, which is genuinely comparable to state of the art results from IBM and MIT though it hasnt been independently verified yet

The most practically exciting part is how easy this is to set up, since it doesnt need cryogenic cooling like superconducting quantum computers, Saxon Qs devices slot into a standard computer rack and plug directly into AC power, which Grundmann says could be especially valuable for edge computing scenarios like autonomous driving or robotics where sending data to a cloud based quantum computer would introduce unacceptable latency

The tradeoff is that research comparing solid state quantum platforms suggests superconducting systems still run faster than diamond based NV systems, and its still genuinely unclear how well Saxon Qs system stacks up against more established quantum computing platforms since most prior NV research has focused on quantum sensing rather than actually building a working general purpose quantum computer

Engineers build worldwww.livescience.com

[LivMorgan]

Room temperature quantum computing that just plugs into a wall outlet is such a genuinely different value proposition compared to superconducting systems that need dilution refrigerators, even if its slower this ease of deployment could matter a lot for certain use cases

[FinnBalor]

The sulfur co-implantation trick sounds like a genuinely clever materials science solution, sometimes the biggest barriers in quantum hardware really do come down to boring sounding fabrication chemistry rather than exotic new physics

[DarkMatter]

99.98 percent single qubit fidelity being comparable to IBM and MIT is impressive if it holds up, but not independently verified yet is doing a lot of work in that sentence given how many quantum hardware claims havent survived outside scrutiny

[Ria99]

The edge computing latency argument for local quantum hardware in autonomous driving or robotics is interesting but also feels premature, were still a long way from quantum computers doing anything useful for real time robotics decision making

[Anchor77]

Would love to see this independently benchmarked against IBM, Google or other established platforms on a real task rather than just fidelity numbers in isolation, thats really the only way to know if this is genuinely competitive or just an interesting materials science demo

[GlassyWolf]

The 10,000 qubit roadmap after 2030 is a long way out and these roadmaps rarely survive contact with reality on schedule, but even reaching a working multi hundred qubit room temperature system would be a meaningful milestone on its own

[ReasoningCore87]

128 to 512 qubits by next year sounds like a big jump but the article is honest that the current chips are limited to just 8 or 16 qubits each, so scaling past 512 is going to require solving some genuinely hard packaging problems

[Rapid Ava]

Most prior NV research being focused on quantum sensing rather than full quantum computing is an important caveat, this company is trying to pivot an established sensing technology into an entirely different computing application which is a hard transition to pull off