Recent advancements in optical computing have demonstrated the potential of using light to process data, offering a more energy-efficient alternative to traditional electronic systems. In a notable development, an optical computer successfully played the classic arcade game Pac-Man, showcasing its capability to handle complex tasks. This achievement underscores the promise of photonic processors in reducing the substantial energy demands associated with artificial intelligence (AI) computations.
Silicon Valley startup Lightmatter has unveiled a new type of computer chip that utilizes beams of light instead of electrical signals to accelerate AI computations while significantly reducing energy consumption. This photonic chip, developed in partnership with GlobalFoundries, addresses the limitations of traditional chip designs by employing light for data transmission and processing, thereby overcoming the scaling challenges of electronic transistors. The chip’s enhanced precision allows it to match conventional processors in certain AI tasks, marking a critical step toward the future of computing.
News & Related Coverage
Context & Background
Optical computing, or photonic computing, involves using light waves produced by lasers or other sources for data processing, storage, and communication. This approach offers higher bandwidth and energy efficiency compared to traditional electronic computing, which relies on electrons. The fundamental building block of modern electronic computers is the transistor, and replacing electronic components with optical ones requires the development of equivalent optical transistors. Recent research has focused on creating optical logic gates and processors that can perform computations directly with light, aiming to overcome the limitations of electronic systems.
In October 2024, an international team of electrical engineers developed a new method for photonic in-memory computing, achieving a faster, more efficient, and robust memory cell. This breakthrough addressed current limitations by combining non-volatility, multibit storage, high switching speed, low switching energy, and high endurance in a single platform. The researchers demonstrated that materials previously used for static optical applications could serve as a platform for high-performance photonic memory, marking a significant step toward scalable optical computing architectures.
In This Story
Lightmatter
Lightmatter is a Silicon Valley startup specializing in photonic computing technologies. The company focuses on developing computer chips that use light for data processing, aiming to enhance the speed and energy efficiency of artificial intelligence computations.
Nick Harris
Nick Harris is the CEO of Lightmatter. He has been instrumental in advancing the development of photonic processors and advocating for their potential to revolutionize computing by overcoming the limitations of traditional electronic systems.
GlobalFoundries
GlobalFoundries is a multinational semiconductor contract manufacturing and design company. It collaborates with various technology firms, including Lightmatter, to produce advanced semiconductor solutions.
Lightelligence
Lightelligence is a technology company based in Singapore that focuses on developing optical computing solutions. The company has demonstrated the potential of optical chips to solve complex computational problems more efficiently than traditional computers.
