Martin Croome, vp of business development, said: “GAP8 is aimed at battery powered devices performing content understanding and control applications. Examples include keyword spotting, beam forming and speech analysis. It could also be used for vibration analysis and face detection.”
The part features eight RISC-V cores combined with a convolution hardware accelerator. It also has a separate core with an independent frequency and voltage domain. Croome said: “One side of the chip looks like an MCU; the other side is the RISC-V cluster, which shares instruction caches.”
Interestingly, Greenwaves has developed the part on an apparent shoestring, having only raised €3.1m in August 2017. “It’s a small amount of money for a start up,” Croome noted, “but we’re using open source components. We couldn’t have used anything other than RISC-V cores because that would have needed more investment.”
Amongst the open source technology integrated into GAP8 is the Parallel Ultra Low Power (PULP) computing open-source platform developed at the University of Bologna and ETH Zurich. PULP is said to provide GAP8 with design maturity, an open source software developer community and a full tool chain.
Measuring 7 x 7mm and manufactured on TSMC’s 55nm LP process, the chip is said to consume 20 times less power than comparable devices. Croome said: “It offers several GOPs of processing power while consuming mW. The device runs the PULP OS and mBed, but we are looking at supporting FreeRTOS.”
According to Greenwaves, the device can run at frequencies of up to 133MHz from a 1V supply, with 250MHz possible from a 1.2V source.
A GAP8 hardware development kit, scheduled for release in April, will include GAPDUINO, pictured, an Arduino compatible board, and a software development kit.