When it comes to quantum computing, the number of qubits a machine has is not the only factor that determines its success. However, Intel believes that its approach, which closely resembles traditional computers, will ultimately lead to breakthroughs and highly functional qubits.
Intel is currently behind some competitors in the quantum computing race. Nevertheless, it aims to surpass them with its quantum computer processors, which have the potential to revolutionize fields such as battery and solar device creation, cost optimization, inexpensive fertilizer production, and even waterproof clothing. Quantum computers also hold promise for advancing artificial intelligence.
Quantum computing is built on the principles of ultrasmall physics. While traditional computers store information in bits that can be either zeros or ones, qubits, the fundamental building blocks of quantum computers, can exist in a unique combination of both states, thanks to a phenomenon called superposition. By combining multiple qubits, these computers can perform complex calculations that surpass the capabilities of traditional computing.
However, qubits are highly sensitive to external disturbances that can disrupt their calculations. To mitigate this issue, one approach is to group multiple qubits together into a single error-correcting element that maintains stability. However, error correction requires even more qubits.
“To achieve optimal computing performance, you need to scale up to millions of qubits and millions of error-correcting elements,” said Intel Chief Technology Officer Greg Lavender at the Intel Innovation conference.
While it is too early to determine the outcome, analyst James Sanders from CCS Insight believes that Intel’s approach holds promise. He remarked, “Intel’s focus on developing qubits using silicon could be a winning strategy. Whether it will make Intel a market leader remains to be seen.”
Prioritizing Quantum Quality
Intel’s competitors currently have quantum machines with a higher number of qubits, such as Intel’s Tunnel Falls quantum processor, which contains more than 12 qubits. However, Intel is not solely focused on quantity; it prioritizes improving the quality of its qubits.
“We are working on another project,” said Intel Labs Director Rich Uhlig, without disclosing specifics. “For us, it’s less about numbers and more about quality.”
At present, the Tunnel Falls processor hosts 24,000 qubits on a 300mm silicon wafer. However, these qubits need to be upgraded to enhance their reliability, communication capabilities, and error correction, according to Intel CEO Pat Gelsinger.
Intel is also developing the Horse Ridge processor, a technology that aims to better control qubits. Overcoming the challenges of quantum processors, which require extremely low temperatures and generate excess heat, poses a significant hurdle.
Testing quantum hardware is also a time-consuming process, as it requires hours to cool the system enough for quantum computing to work. To expedite hardware development, Intel has created a device that can test thousands of processors during the heating phase.
Diverse Approaches to Qubits
Traditional computers rely on transistors embedded in silicon crystals to perform computations. In contrast, companies exploring quantum computing are pursuing various approaches. It remains unclear which approach will dominate, or whether multiple approaches will coexist.
The circuit board that houses Intel’s Tunnel Falls quantum processor is about the size of an adult’s hand.
IBM and Google predominantly focus on superconducting qubits, which are extremely cold tiny circuits. IonQ utilizes ion trap technology to trap electrically charged atoms, allowing for slower but more reliable interactions. Other companies are exploring electrically neutral atoms or particles called photons.
Instead of pursuing superconducting qubits like many competitors, Intel has opted for spin qubits that resemble its traditional microprocessors. These spin qubits, which utilize electrons in silicon chips, have the potential to rival the computing power offered by other technologies.
“I hope that by 2030, there will be an alternative to transmon [superconducting] or ion trap qubits that surpasses current quantum computing capabilities,” stated James Sanders.
