The quantum computing narrative is definitely ramping up, and in the midst of it, IonQ's CEO claims that the firm's quantum chips will 'wipe out' the competition from NVIDIA and AMD.
IonQ Plans To Scale Up Quantum Chips To 10,000 Qubits By 2027, But Could They Replace Traditional GPUs?
Well, there have always been questions about the future of computing, or whether GPU manufacturers like NVIDIA or AMD could rely on architectural advancements and even Moore's Law to scale up performance. One prospect that the industry sees is the use of quantum computers for applications which require processing power, and with that, several QC firms are emerging from the 'depths of the market'. And now, one of the more popular firms, IonQ, claims that they could beat classical GPUs in several applications, even if they had the "known age of the universe".
Of course, IonQ's CEO Niccolo de Masi's claims of beating classical GPUs are debatable, but interestingly, he also stated that NVIDIA's Blackwell architecture could become outdated by 2027, courtesy of the firm's upcoming 10,000-qubit quantum chip. Let's open up these claims to see whether QC firms could actually compete with NVIDIA/AMD. At their current state, quantum computers are known to be best towards specific application such as optimization, chemistry, cryptography, but in classical compute, the situation is different.
IonQ's CEO was talking with Bloomberg on the acquisition of Oxford Ionics, and the move has accelerated the company's roadmap to the point where IonQ expects 10,000 physical qubits on a single chip by 2027. If you are unaware of the fundamentals behind a quantum chip, let me dedicated a few words. Quantum chips are equipped with physical and logical qubits, with the earlier one acting as superconducting circuits to hold encoded data, but logical qubits are actually responsible for error-corrected quantum algorithms, that developers leverage for computation.
Even if we reach two million qubits by 2030, that will let us solve problems that classical GPUs couldn’t handle, even if they had the age of the universe. They [Blackwell chips] will look outdated long before then — by 2027, with 10,000 qubits developed in partnership with our new friends at Oxford Ionics, and honestly, that will wipe the floor with every supercomputer that exists on Earth.
Now, comparing quantum chips with classical GPUs is like comparing a Formula 1 car to a cargo truck, since the latter excels in parallel workloads, which come with predictable, exact results, while qubit chips are designed to target problems that grow exponentially harder for classical machines. The only way companies like IonQ could compete with NVIDIA or AMD GPUs is by ramping up logical qubits, which to say the least, remains the greatest challenge today.
About the author: Muhammad Zuhair is a hardware and technology reporter for Wccftech, specializing in the semiconductor industry and the complex interplay between technology, manufacturing, and geopolitics. His coverage focuses on the corporate strategies and technological roadmaps of industry giants like TSMC, NVIDIA, Samsung, and Intel. Zuhair's expertise lies in deconstructing complex topics such as fabrication nodes (e.g., 2nm process), the economic impact of policies like the CHIPS Act, and the strategic development of AI infrastructure from NVIDIA, AMD and Intel.
Follow Wccftech on Google to get more of our news coverage in your feeds.
