A few weeks back, I wrote about the need for machine learning at the edge and what big chip firms are doing to address the challenge. Even as Intel, ARM, and others invest in new architectures, startups are also attempting to innovate with new platforms.
GreenWaves Technologies, based in France, is one such company. It has built a machine learning chip that offers multiple cores and low-power machine learning at the edge. The chip is called the GAP8 application processor. GreenWaves CEO Loic Lietar says the GAP8 processor can offer always-on face detection with a few milliwatts of power, indoor people counting with years of battery life, and sub-$15 machine vision or voice control for consumer applications.
But what’s really fascinating about GreenWaves is that it has built this chip with so little in funding. The company has raised 3.1 million euros ($3.8 million) so far, much less than a traditional semiconductor startup. And yet it’s gotten all the way to producing chips based on its design, with a development board coming in April. The reason GreenWaves could do so much with so little is because it’s building on a new open-source hardware architecture called RISC-V.
RISC-V was created eight years ago as a project at UC Berkeley aimed at building low-power chips that use minimal instruction sets. An instruction set governs how software talks to the actual computing elements on the chip, and anyone can use it to build their own designs. By comparison, Intel keeps its x86 instruction set to itself (and AMD), while the instruction sets of ARM and MIPS are licensed out for millions of dollars. (For more on this dynamic, here’s a good article.)
GreenWaves’ Lietar says the cost of building a new chip using ARM’s instruction set would start with a $15 million license and escalate from there. But with RISC-V his engineers could simply download the code and get going. Of course, to design a chip at this level still requires sophisticated engineers experienced in building processors. However, GreenWave’s founders hail from ST Microelectronics and have lots of chip design experience.
I’m excited by GreenWaves’ chips, but I’m even more excited that in an era when we are going to need different types of processors designed for low-power, edge computing jobs, there’s potentially a way to innovate in silicon at a lower cost. Because in my opinion, silicon is where your innovation starts. The capability has to be there in hardware before you can do new things with software.