artificial intelligence

Posted on by Debashis Das - No Comments on Infineon launches PSoC Edge Cortex-M55/M33 microcontrollers for enhanced AI and ML applications

Infineon launches PSoC Edge Cortex-M55/M33 microcontrollers for enhanced AI and ML applications

The PSoC Edge microcontrollers integrate the Arm Cortex-M55 core with Helium DSP and the Ethos U55 NPU unit for advanced AI tasks. It also includes a power-efficient Arm Cortex-M33 core paired with an NNLite(DSP/NPU) for simpler AI tasks.

Infineon’s new PSoC Edge series of microcontrollers integrates the Arm Cortex-M55 core with Helium DSP and the Ethos U55 NPU unit for advanced AI tasks. It also includes a power-efficient Arm Cortex-M33 core paired with an NNLite(DSP/NPU) for simpler AI tasks. This setup allows the device to be more efficient in varying load conditions. Infineon’s PSoC lineup is a configurable microcontroller powered by Arm Cortex-M4, Cortex-M3, or Cortex-M0+ cores. The main USP (Unique Selling Point) of this lineup is its configurable digital and analog components. This feature makes them somewhat similar to an FPGA, but an FPGA is far more difficult to program than these microcontrollers and they are also very power-hungry. The device is equipped with advanced HMI and “Always-on” capability. “Always-on” is a feature of this device to constantly monitors and responds to signals automatically, making it suitable for smart homes, security, wearables, robotics, and many more. Key […]

Posted on by Debashis Das - No Comments on Firefly AIO-3588Q 8K AI Board Features Octa-Core CPU and 6 TOPS NPU

Firefly AIO-3588Q 8K AI Board Features Octa-Core CPU and 6 TOPS NPU

AIO 3588Q 8K AI Board

Firefly has recently launched the AIO-3588Q 8K AI board powered by a powerful Rockchip RK3588 SoC, up to 32GB of RAM, 8K video encoding-decoding, Gigabit Ethernet, Wi-Fi 6, 5G/4G, and multiple display support. With all these features this board is designed to target the industrial, commercial, and automotive markets. Previously, we have covered many SBCs and SoMs designed by Firefly such as the similarly named Firefly AIO-3588SG; feel free to check those out if you’re interested. Features and specifications of the AIO-3588Q 8K AI Board: SoC – RockChip RK3588: CPU – Octa-core 64-bit (4x Cortex-A76+4x Cortex-A55), 8nm process, up to 2.4GHz. GPU – ARM Mali-G610 MP4, supports OpenGL ES3.2/OpenCL 2.2/Vulkan1.1, 450 GFLOPS. NPU – Up to 6 TOPS, supports INT4/INT8/INT16, compatible with TensorFlow/MXNet/PyTorch/Caffe. ISP – Integrated 48MP ISP with HDR & 3DNR. VPU – Supports 8K @ 60fps and 4K @ 60fps video decoding/encoding, including H.265/VP9/AVS2, and H.264 AVC/MVC. RAM […]

Posted on by Jean-Luc Aufranc (CNXSoft) - 5 Comments on COM Express Type 6 Compact module features 14th gen Intel Core Ultra “Meteor Lake” processor

COM Express Type 6 Compact module features 14th gen Intel Core Ultra “Meteor Lake” processor

Intel Core Ultra COM Express

ADLINK cExpress-MTL is a COM Express Type 6 Compact module based on the just-announced 14th gen Intel Core Ultra “Meteor Lake” processor family with up to fourteen CPU cores in 6P+8E configuration, eight Xe-cores (128 EUs), and an NPU (11pTOPS/8.2eTOPS) in a 15 or 28W TDP thermal configuration delivering up to 1.9x the GPU performance of the previous generation (Raptor Lake). The power consumption will also be lower thanks to Intel Core Ultra’s new Low-Power E-cores that are 30 to 50% more efficient than the E-cores in 13th gen Intel Core processors, and the faster GPU and built-in NPU will enable hardware-accelerated AV1 encoding/decoding and various devices leveraging AI such as portable medical ultrasound devices, industrial automation, autonomous driving, AI robots, and more. cExpress-MTL specifications: Meteor Lake-H/U SoC (one of the other) Intel Core Ultra 7 MS3 165H 16-core (6P+8E+2LPE) processor @ 1.4 / 5.0 GHz with 24MB cache, Intel […]

Posted on by Jean-Luc Aufranc (CNXSoft) - No Comments on Renesas RA8D1 Arm Cortex-M85 SoC features LCD display interfaces, 2D graphics accelerator

Renesas RA8D1 Arm Cortex-M85 SoC features LCD display interfaces, 2D graphics accelerator

Renesas RA8D1

Renesas RA8D1 is a new Arm Cortex-M85 microcontroller with graphics capabilities such as a 2D graphics accelerator and MIPI DSI and parallel RGB interfaces to connect an LCD that will make the chip suitable for HMI applications. Renesas introduced the world’s first Arm Cortex-M85 microcontroller with the RA8M1 just a few weeks ago, but the MCU has limited multimedia capabilities with just a 16-bit Capture Engine Unit (CEU) interface to connect a camera. The second member of the Renesas RA8 family fills this void with the RA8D1 microcontroller adding an LCD controller and a 2D graphics drawing engine on top of the CEU camera interface. Renesas RA8D1 specifications: MCU core – Arm Cortex-M85 clocked at 480 MHz with Helium MVE (M-Profile Vector Extension) with 32KB I/D caches, 12KB data flash Memory & Storage 1MB SRAM with TCM (128KB) 1MB to 2MB Flash memory 32-bit external SDRAM interface Display interfaces and […]

Posted on by Jean-Luc Aufranc (CNXSoft) - No Comments on Mixtile Core 3588E development kit review – Part 1: Unboxing and first boot

Mixtile Core 3588E development kit review – Part 1: Unboxing and first boot

Mixtile Core 3588E development kit review

Mixtile Core 3588E is a Rockchip RK3588 SO-DIMM system-on-module with up to 32GB LPDDR4 and 256GB eMMC flash that follows NVIDIA Jetson Nano, TX2 NX, Xavier NX, and Orin Nano form factor and should work with any board made for Jetson SO-DIMM modules. We first covered the module last April, but Mixtile is now shipping their Rockchip RK3588 module and sent us a development kit with the Mixtile Core 3588 SoM fitted with a fansink and a carrier board that looks quite similar to the NVIDIA Jetson Nano developer kit we reviewed a few years ago, although its design is closer to the new B01 developer kit as we’ll see below.  We’ll start the review with an unboxing and first boot, before testing it in more detail with Ubuntu 22.04 in the second part of the review. Mixtile Core 3588E development kit unboxing We received the development kit fully preassembled. […]

Posted on by Jean-Luc Aufranc (CNXSoft) - No Comments on Tokay Lite – A battery-powered no-code AI camera based on ESP32-S3 WiSoC (Crowdfunding)

Tokay Lite – A battery-powered no-code AI camera based on ESP32-S3 WiSoC (Crowdfunding)

Tokay Lite ESP32-S3 no-code AI camera

Maxlab’s Tokay Lite is an OHSWA-certified AI camera based on ESP32-S3 WiFI and Bluetooth SoC that can be used for computer vision (e.g. facial recognition & detection) and robotics applications without the need to know programming languages since a web interface is used for configuration. The WiFi and Bluetooth AI camera also features night vision with four IR LEDs, an IR cut filter, light and PIR motion sensors, a 20-pin expansion connector with SPI and UART, support for an external RTC, and can take power from USB-C or a LiPo battery. Tokay Lite specifications: Wireless module ESP32-S3-WROOM-1 MCU – ESP32-S3 dual-core LX7 microprocessor @ up to 240 MHz with Vector extension for machine learning, 512 KB SRAM Memory – 8MB PSRAM Storage – 8MB SPI flash Connectivity – WiFi 4 and Bluetooth 5 with LE/Mesh PCB antenna Certifications – FCC/CE certification Camera OV2640 camera sensor (replaceable) via DVP interface Image Capabilities: […]

Posted on by Jean-Luc Aufranc (CNXSoft) - No Comments on Robustel introduces EG5101 and EG5200 Debian 11 industrial IoT gateways with 4G LTE cellular connectivity

Robustel introduces EG5101 and EG5200 Debian 11 industrial IoT gateways with 4G LTE cellular connectivity

Robustel EG5101 EG5200 Debian 11 industrial gateways

Robustel introduced the EG5101 and EG5200 industrial IoT gateways running Debian 11 a few weeks ago, both with 4G LTE cellular connectivity, but the NXP i.MX 6ULL-based EG5101 has a more compact design, while the NXP i.MX 8M Plus-powered EG5200 provides more I/Os and resources. Expanding the earlier EG5100, EG5120, and LG5100 models from the company, the two edge computing gateways target Industry 4.0 and smart infrastructure applications and enable real-time analytics at the edge with lower latency than running workloads in the cloud. Robustel EG5101 and EG5200 SoC/Memory/Storage EG5101 NXP i.MX 6ULL Cortex-A7 32-bit processor @ 792 MHz 512MB to 1024MB DDR3 8GB eMMC flash EG5200 NXP i.MX 8M Plus quad-core Cortex-A53 64-bit processor @ 1.6 GHz with 2.3 TOPS NPU 4GB DDR4 32GB eMMC flash, microSD card socket Connectivity EG5101 1x 10/100 Mbps Ethernet port 4G LTE with 1x SMA-K antenna connector, 2x Mini SIM (2FF) EG5200 5x […]

Posted on by Jean-Luc Aufranc (CNXSoft) - 13 Comments on Orange Pi AIPro SBC features a 20 TOPS Huawei Ascend AI SoC

Orange Pi AIPro SBC features a 20 TOPS Huawei Ascend AI SoC

Orange Pi Huawei Ascent SBC

Orange Pi AIPro is a new single board computer for AI applications that features a new (and unnamed) Huawei Ascend AI quad-core 64-bit processor delivering up to 20 TOPS (INT8) or 8 TOPS (FP16) of AI inference performance. The SBC comes with up to 16GB LPDDR4X and a 512Mbit SPI flash but also supports other storage options such as a microSD card, an eMMC flash module, and/or an M.2 NVMe or SATA SSD. The board also features two HDMI 2.0 ports, one MIPI DSI connector, and an AV port for video output, two MIPI CSI camera interfaces, Gigabit Ethernet and WiFi 5 connectivity, a few USB ports, and a 40-pin GPIO header for expansion. Orange Pi AIPro specifications: SoC – Huawei Ascend quad-core 64-bit (I’d assume RISC-V) processor delivering up to 20 TOPS (INT8) or 8TOPS (FP16) AI performance and equipped with an unnamed 3D GPU System Memory – 8GB […]