Chipmakers such as Nvidia Corporation (NVDA), Intel Corporation (INTC), Advanced Micro Devices, Inc. (AMD) and ARM Holdings could face significant headwinds as their general-purpose processors lose market share to hardware designed for specific tasks, according to Barron's. Google, which has developed an innovative chip called a Tensor Processing Unit (TPU), could present a notable threat to these better established chip producers. (For more, see also: AMD vs. Nvidia: Who Dominates GPUs?)
Google, which is owned by Alphabet Inc (GOOGL), announced the second generation of these TPUs in May, TechCrunch reported. Thus far, these chips have outperformed similar ones produced by Nvidia and Intel when used for specific tasks such as machine learning. Due to their efficient nature, they can be harnessed for machine learning models, which could correctly determine what kind of object a specific photograph contains.
The TPU is more specialized than general-purpose chips, making it better-designed to complete specific tasks, according to Barron's. When compared a similar Intel microprocessor, the TPU has 3.5 times as much memory. "We threw out a lot of stuff that was not needed," says David Patterson, a University of California computer scientist who also works for Google as a distinguished engineer. "Instead of the Honda for everyone, we are making these Formula One race cars for some things." (For more, see also: Why AMD Is Intel’s Only Competitor.)
The TPU also took less time to design, Barron's reported. While Intel processors can take years to develop, Google's new chip went from the blueprint phase to completion in only 15 months, Patterson told Barron's. Going forward, shorter production cycles could become the norm, he maintained. "We are at a paradigm shift in computing architecture," said Patterson.
Not everyone is convinced of Patterson's point of view, as some think that designing processors for specific tasks will confine them to niche functions, according to Barron's. In the end, the markets will determine who was right in their predictions, Patterson told Barron's.