sdk

Posted on by Jean-Luc Aufranc (CNXSoft) - 6 Comments on EdgeCortix SAKURA-II Edge AI accelerator deliver up to 60 TOPS in an 8W power envelope

EdgeCortix SAKURA-II Edge AI accelerator deliver up to 60 TOPS in an 8W power envelope

SAKURA-II M.2 and PCIe Edge AI accelerators

EdgeCortix has just announced its SAKURA-II Edge AI accelerator with its second-generation Dynamic Neural Accelerator (DNA) architecture delivering up to 60 TOPS (INT8) in an 8Watts power envelope and suitable to run complex generative AI tasks such as Large Language Models (LLMs), Large Vision Models (LVMs), and multi-modal transformer-based applications at the edge. Besides the AI accelerator itself, the company designed a range of M.2 modules and PCIe cards with one or two SAKURA-II chips delivering up to 120 TOPS with INT8, 60 TFLOPS with BF16 to enable generative AI in legacy hardware with a spare M.2 2280 socket or PCIe x8/x16 slot. SAKURA-II Edge AI accelerator SAKURA-II key specifications: Neural Processing Engine – DNA-II second-generation Dynamic Neural Accelerator (DNA) architecture Performance 60 TOPS (INT8) 30 TFLOPS (BF16) DRAM – Dual 64-bit LPDDR4x (8GB,16GB, or 32GB on board) DRAM Bandwidth – 68 GB/sec On-chip SRAM – 20MB Compute Efficiency – […]

Posted on by Debashis Das - 1 Comment on Avnet AI Vision Development Kit features Qualcomm QCS6490 SoC, dual camera, GbE, and USB-C PD

Avnet AI Vision Development Kit features Qualcomm QCS6490 SoC, dual camera, GbE, and USB-C PD

QCS6490 Vision AI Development Kit

Just last month at Embedded World 2024, Qualcomm announced its RB3 Gen 2 Platform based on the QCS6490 processor with Cortex-A78 and A55 processing cores and 12 TOPS of AI power. Building on this, Avnet has recently launched the Avnet AI Vision Development Kit, also based on the QCS6490 SoC. The kit includes a dual camera setup, Gigabit Ethernet connectivity, USB-C Power Delivery, and a host of other features for applications like inventory and asset monitoring, drone/UAV/other mobile vision-AI edge compute applications, and multi-camera security systems with recognition. Previously we have written about similar AI dev kits including Allwinner V853 100ASK-V853-Pro, Sipeed Maix-III devkit, RZBoard V2L, and many other AI vision development boards feel free to check those out if you are interested in the topic. Avnet AI Vision Development Kit specifications SM2S-QCS6490 SMARC Compute Module: CPU – 4x Arm Cortex-A78 (up to 2.7 GHz), 4x Arm Cortex-A55 (up to […]

Posted on by Jean-Luc Aufranc (CNXSoft) - 10 Comments on Matter 1.3 specification adds support for water and energy management, electric vehicle chargers, and various household appliances

Matter 1.3 specification adds support for water and energy management, electric vehicle chargers, and various household appliances

Matter 1.3 Specification

The Connectivity Standard Alliance (CSA) has just announced the release of the Matter 1.3 specification and SDK with energy reporting, support for water and energy management devices, electric vehicle chargers, several new “major appliances”, namely various kitchen appliances and laundry dryers, and various other features. As a reminder the Matter protocol was initially introduced several years ago under the name Project CHIP to improve the interoperability of Smart Home devices from various vendors, so for example, users could connect a Matter-compatible Philips Hue light bulb to a Samsung gateway, or a white-brand Matter sensor with Google Home, etc… Matter started to pick last year with several products launched, and Paisit notably reviewed the MINI Extreme Wi-Fi Smart Switch (MINIR4M), the first Matter device from SONOFF, last October. Matter 1.3 adds various new capabilities and devices. Matter 1.3 highlights: Support for Water and Energy Management Devices Energy Management – Matter 1.3 […]

Posted on by Dennis Mwihia - 2 Comments on MemryX MX3 edge AI accelerator delivers up to 5 TOPS, is offered in die, package, and M.2 and mPCIe modules

MemryX MX3 edge AI accelerator delivers up to 5 TOPS, is offered in die, package, and M.2 and mPCIe modules

MemryX MX3 EVB

Jean-Luc noted the MemryX MX3 edge AI accelerator module while covering the DeGirum ORCA M.2 and USB Edge AI accelerators last month, so today, we’ll have a look at this AI chip and corresponding modules that run computer vision neural networks using common frameworks such as TensorFlow, TensorFlow Lite, ONNX, PyTorch, and Keras. MemryX MX3 Specifications MemryX hasn’t disclosed much performance stats about this chip. All we know is it offers more than 5 TFLOPs. The listed specifications include: Bfloat16 activations Batch = 1 Weights: 4, 8, and 16-bit ~10M parameters stored on-die Host interfaces – PCIe Gen 3 I/O and/or USB 2.0/3.x Power consumption – ~1.0W 1-click compilation for the MX-SDK when mapping neural networks that have multiple layers Under the hood, the MX3 features MemryX Compute Engines (MCE) which are tightly coupled with at-memory computing. This design creates a native, proprietary dataflow architecture that utilizes up to 70% […]

Posted on by Jean-Luc Aufranc (CNXSoft) - 12 Comments on Sipeed MaixCAM is a RISC-V AI camera devkit with up to 5MP camera, 2.3-inch color touchscreen display, GPIOs

Sipeed MaixCAM is a RISC-V AI camera devkit with up to 5MP camera, 2.3-inch color touchscreen display, GPIOs

Sipeed MaixCAM

Sipeed MaixCAM is an AI camera based on SOPHGO SG2002 RISC-V (and Arm, and 8051) SoC with a 1 TOPS NPU that takes up to 5MP camera modules and comes with a 2.3-inch color touchscreen display. The development kit also comes with WiFi 6 and BLE 5.4 connectivity, optional Ethernet, audio input and output ports, a USB Type-C port, and two 14-pin GPIO headers for expansion that makes it suitable for a range of computer vision, Smart audio, and AIoT applications. Sipeed MaixCAM specifications: SoC – SOPHGO SG2002 CPU 1 GHz RISC-V C906 processor or Arm Cortex-A53 core (selectable at boot) running Linux 700 MHz RISC-V C906 core running an RTOS 25 to 300 MHz low-power 8051 processor NPU – 1 TOPS @ INT8 with support for models such as Mobilenetv2, YOLOv5, YOLOv8, etc… Video Codec – H.264, H.265, MJPEG hardware encoding and decoding up to 2K @ 30fps Memory […]

Posted on by Jean-Luc Aufranc (CNXSoft) - No Comments on Yocto Project 5.0 “Scarthgap” released with Linux 6.6 and plenty of changes

Yocto Project 5.0 “Scarthgap” released with Linux 6.6 and plenty of changes

Yocto Project 5.0

The Yocto Project 5.0 codenamed “Scarthgap” has just been released with Linux 6.6, glibc 2.39, LLVM 18.1, and over 300 other recipe upgrades. As a result of the release, the developers have made it available for download (bz2 tarball). The Yocto Project, or Yocto for shorts, is a popular framework used to create custom embedded Linux distributions, and we’ve played with it over the year showing how to create a minimal image for the Raspberry Pi, and last year, we used it again when reviewing two industrial development boards, namely the VOIPAC IMX8M and ADLINK i-Pi SMARC 1200. Yocto is quite a powerful framework/build system with plenty of options that make it highly customizable, but the learning curve is fairly steep. Some other changes in Yocto Project 5.0 include: New variables: CVE_DB_INCR_UPDATE_AGE_THRES: Configure the maximum age of the internal CVE database for incremental update (instead of a full redownload). RPMBUILD_EXTRA_PARAMS: […]

Posted on by Debashis Das - No Comments on Synaptics Astra platform takes SL1620, SL1640, or SL1680 Arm CPU module for Edge AI applications

Synaptics Astra platform takes SL1620, SL1640, or SL1680 Arm CPU module for Edge AI applications

Astra Machina Foundation Series Dev Board

Synaptics has unveiled its new Astra platform with a range of SoC and a development kit for edge AI applications. These new processors and a supporting development kit are built to provide out-of-the-box AI capabilities for IoT devices, reducing reliance on cloud-based AI. This new Synaptics Astra Platform is built around three main SoCs. The SL1680 is built for multi-modal IoT applications and features a quad-core Arm Cortex-A73 CPU, dedicated 7.9 TOPS NPU, and 4K video. The SL1640 is a cost and power-optimized SoC with a quad-core Cortex-A55 CPU and 1.6+ TOPS NPU. Finally, the SL1620 is a graphics and AI accelerator with a quad-core Cortex-A55 CPU and dual-core Imagination BXE-2-32 GPU but does not feature an NPU. The development kit features a module design where the new swappable compute modules allow flexible configurations. The devkit will support open Yocto Linux distribution and Synaptics AI toolkit for quick AI integration. […]

Posted on by Jean-Luc Aufranc (CNXSoft) - 6 Comments on Qualcomm RB3 Gen 2 Platform with Qualcomm QCS6490 AI SoC targets robotics, IoT and embedded applications

Qualcomm RB3 Gen 2 Platform with Qualcomm QCS6490 AI SoC targets robotics, IoT and embedded applications

Qualcomm RB3 Gen 2 Platform Vision Kit

Qualcomm had two main announcements at Embedded World 2024: the ultra-low-power Qualcomm QCC730 WiFi microcontroller for battery-powered IoT devices and the Qualcomm RB3 Gen 2 Platform hardware and software solution designed for IoT and embedded applications based on the Qualcomm QCS6490 processor that we’re going to cover today. The kit is comprised of a QCS6490 octa-core Cortex-A78/A55 system-on-module with 12 TOPS of AI performance, 6GB RAM, and 128GB UFS flash connected to the 96Boards-compliant Qualcomm RBx development mainboard through interposer, as well as optional cameras, microphone array, and sensors. Qualcomm QCS6490/QCM6490 IoT processor Specifications: CPU – Octa-core Kryo 670 with 1x Gold Plus core (Cortex-A78) @ 2.7 GHz, 3x Gold cores (Cortex-A78) @ 2.4 GHz, 4x Silver cores (Cortex-A55) @ up to 1.9 GHz GPU – Adreno 643L GPU @ 812 MHz with support for Open GL ES 3.2, Open CL 2.0, Vulkan 1.x, DX FL 12 DSP – Hexagon […]

EmbeddedTS embedded systems design