The computing chips that power artificial intelligence consume a lot of electricity. On Wednesday, the world’s biggest manufacturer of those chips showed off a new strategy to make them more energy efficient: using AI-powered software to design them.
Advertisement
At a conference in Silicon Valley, Taiwan Semiconductor Manufacturing Company (TSMC), the contract manufacturer that fabricates chips for Nvidia, showed off a range of ways that it was hoping to boost the energy efficiency of AI computing chips by about 10 times.
Nvidia’s current flagship AI servers, for example, can consume as much as 1,200 watts during demanding tasks, which would be the equivalent of the power used by 1,000 US homes if run continuously.
The gains TSMC is hoping to achieve come from a new generation of chip designs in which multiple “chiplets” – smaller pieces of full computing chips – using different technologies are packaged together to make one computing package.
But to make use of those technologies, the firms that design chips are increasingly relying on AI-powered software from providers such as Cadence Design Systems and Synopsys, both of which rolled out new products on Wednesday that had been developed in close coordination with TSMC.
Advertisement
For some of the complex tasks in designing chips, the tools from TSMC’s software partners found better solutions than TSMC’s own human engineers – and did so much faster.