Biggish's Blue's huge NorthPole processor chip eliminates the need to frequently access external memory, and so performs tasks such as image recognition faster than existing architectures do -- while consuming vastly less power.

University of Paris-Saclay nanoelectronics researcher Damien Querlioz wrote in Science that the chip proves computing and memory can be integrated on a large scale.

"Its energy efficiency is just mind-blowing... I feel the paper will shake the common thinking in computer architecture."

NorthPole runs neural networks: multi-layered arrays of simple computational units programmed to recognise patterns in data.

A bottom layer takes in data, such as the pixels in an image; each successive layer detects patterns of increasing complexity and passes information on to the next layer. The top layer produces an output that, for example, can express how likely an image is to contain a cat, a car or other object.