Semiconductor company %Nvidia ($NVDA) has unveiled a new high-powered microchip that’s designed to run %ArtificialIntelligence (AI) services, as well as autonomous vehicles and %Metaverse applications.
At an investor day conference, Nvidia introduced its new H100 AI processor, a successor to the A100 that’s used for training artificial intelligence systems to do things such as translate human speech, recognize what’s in photos, and plot a self-driving car’s route through traffic.
Nvidia says the new chip is its most powerful ever. With the company’s NVLink high-speed communication pathway, customers can now link as many as 256 H100 chips to each other into “one mind-blowing graphics processing unit,” Chief Executive Officer (CEO) Jensen Huang said at the conference.
Nvidia also announced the RTX A5500, a new member of its Ampere series of graphics chips for professionals who need graphics power for three dimensional (3D) tasks such as animation, product design, and visual data processing. That coincides with Nvidia’s expanded Omniverse efforts to sell the tools and cloud computing services needed to build the 3D realms known as the “metaverse.”
The new H100 chips will ship in the third quarter of this year, said CEO Huang.
There’s plenty of competition for the H100, which is composed of a whopping 80 billion transistors that make up its data processing circuitry and is built by Taiwan Semiconductor Manufacturing (TSMC).
Rivals to Nvidia’s newest chip include Intel’s upcoming Ponte Vecchio processor, with more than 100 billion transistors, and a host of AI accelerator chips from start-ups such as Graphcore, SambaNova Systems and Cerebras.
Nvidia also announced that it’s working on a new generation of its Hyperion chips for use in self-driving cars, which are due to arrive in 2026. Nvidia said it expects $11 billion U.S. from car chip sales over the next six years, up from $8 billion U.S. in 2021.