Fullmoonfever
Top 20
This came up as featured in the latest edition and link takes you to the below article which may have been posted before?
www.eenewseurope.com
BrainChip has taped out its latest AKD1500 event based machine learning chip on the low power GlobalFoundries 22nm fully depleted silicon-on-insulator (FD-SOI) technology.
The reference design is a key milestone is part of validating BrainChip’s IP across different processes and foundries. It has also been working with Renesas on integrating the IP, called Akida, into a chip that taped out last month.
For always-on AI at the sensor, Akida can offload feed-forward networks without the use of a CPU, which enables extremely energy efficient intelligent sensors. For other low-power embedded applications, the CPU solely performs sensor fusion processes or runs the application with Akida managing all the AI acceleration to significantly minimize CPU workload and system load, freeing up the MCU for other general-purpose computation.
“This is an important validation milestone in the continuing innovation of our event-based, neuromorphic Akida IP, even if it is fully digital and portable across foundries,” said Anil Mankar, Co-Founder & Chief Development Officer at BrainChip. “The AKD1500 reference chip using GlobalFoundries’ very low-leakage FD SOI platform, showcases the possibilities for intelligent sensors in edge AI.”

BrainChip tapes out its AI chip in 22nm FD SOI
BrianChip has taped out its AKD1500 event based machine learning chip on the low power GlobalFoundries 22nm FD-SOI technology.

BrainChip has taped out its latest AKD1500 event based machine learning chip on the low power GlobalFoundries 22nm fully depleted silicon-on-insulator (FD-SOI) technology.
The reference design is a key milestone is part of validating BrainChip’s IP across different processes and foundries. It has also been working with Renesas on integrating the IP, called Akida, into a chip that taped out last month.
- Renesas tapes out spiking neural network chip
- NASA looks to BrainChip’s spiking neural network chip
- Akida spiking neural processor could head to FDSOI
For always-on AI at the sensor, Akida can offload feed-forward networks without the use of a CPU, which enables extremely energy efficient intelligent sensors. For other low-power embedded applications, the CPU solely performs sensor fusion processes or runs the application with Akida managing all the AI acceleration to significantly minimize CPU workload and system load, freeing up the MCU for other general-purpose computation.
Spiking neural networks
The AKD1500 is designed as an accelerator reference chip, using quad/octal SPI interfaces for MCU modules and PCIe for extensible cards, which assists partners in developing and demonstrating their solutions as a stepping stone to integrating the Akida IP into their production SoCs.“This is an important validation milestone in the continuing innovation of our event-based, neuromorphic Akida IP, even if it is fully digital and portable across foundries,” said Anil Mankar, Co-Founder & Chief Development Officer at BrainChip. “The AKD1500 reference chip using GlobalFoundries’ very low-leakage FD SOI platform, showcases the possibilities for intelligent sensors in edge AI.”