Download a free trial of the SoftNeuro Deep Learning SDK for Intel and Arm targets (Sponsored)

Jetson Xavier Tensorflow Lite vs SoftNeuro

Morpho, a global research & development company established in Japan in 2004 and specialized in imaging technology, is now offering a free trial for the SoftNeuro deep learning SDK working on Intel processors with AVX2 SIMD extensions, 64-bit Arm targets, while also leveraging OpenCL and/or CUDA. Some of the advantages of SoftNeuro are that the framework is easy to use even for those without any knowledge about deep learning, it’s fast thanks to the separation of the layers and their execution patterns, and it can run on several different hardware and OS being cross-platform. SoftNeuro relies on its own storage format (DNN format) to deliver the above advantages. But you can still use models trained with any mainstream deep learning framework. TensorFlow and Keras models can be directly converted to the DNN format, while models from other frameworks can be converted first to the ONNX format and then to the […]

VisionFive V1 RISC-V Linux SBC resurrects BeagleV StarFive single board computer

VisionFive V1 RISC-V SBC

Last summer we reported that BeagleV StarFive RISC-V SBC would not be manufactured, but all was not lost as StarFive would collaborate with Radxa to make a new single board computer based on their JH7100 dual-core 64-bit RISC-V processor. But thanks to a report on Heise and extra photos acquired by CNX Software, we now have more details about the board that mostly comes with the same features as the BeagleV StarFive, but a completely different layout that brings all the main ports to one side of the board. VisionFive V1 specifications: SoC – StarFive JH7100 Vision SoC with: Dual-core Sifive U74 RISC-V processor @ 1.5 GHz with 2MB L2 cache Vision DSP Tensilica-VP6 for computing vision NVDLA Engine 1 core (configuration 2048 MACs @ 800MHz  – 3.5 TOPS) Neural Network Engine (1024MACs @ 500MHz – 1 TOPS) VPU – H.264/H.265 decoder up to 4Kp60, dual-stream decoding up to 4Kp30 […]

Espressif introduces ESP32-S3-BOX AI development kit for online and offline voice applications

ESP32-S3-Box

Espressif Systems has very recently introduced the ESP32-S3-BOX AI voice devkit designed for the development of applications with offline and online voice assistants, and whose design I find similar to the M5Stack Core2 devkit, but the applications will be different. The ESP32-S3-BOX features the latest ESP32-S3 processor with WiFi and BLE connectivity, AI capabilities, as well as a 2.4-inch capacitive touchscreen display, a 2-mic microphone array, a speaker, and I/O connectors with everything housed in a plastic enclosure with a stand. ESP32-S3-BOX specifications: WiSoC – ESP32-S3 dual-core Tensilica LX7 up to 240 MHz with Wi-Fi & Bluetooth 5, AI instructions, 512KB SRAM Memory and Storage – 8MB octal PSRAM and 16MB QSPI flash Display – 2.4-inch capacitive touchscreen display with 320×240 resolution Audio – Dual microphone, speaker USB – 1x USB Type-C port for power and debugging (JTAG/serial) Expansion – 2x Pmod-compatible headers for up to 16x GPIOs Misc Power […]

Jetson EX1 and SUB mini PCs put Xavier NX SoM into metal enclosures

Jetson EX1 mini PC metal enclosure

Seeed Studio has just unveiled another Jetson Xavier NX mini PC with the Jetson EX1, plus some metal enclosures for the Jetson SUB mini PC announced last month with the official NVIDIA Jetson Xavier NX developer kit, a 128GB SSD, and the Re_Computer case comprised of an aluminum frame and a plastic cover. Jetson EX1 mini PC The Jetson EX1 mini PC has a different design with a custom carrier board offering dual Gigabit Ethernet and support for four MIPI CSI cameras, a 256GB NVMe SSD, and an aluminum enclosure designed to help with cooling, albeit the computer is not completely fanless since a 5V fan is included in the design as well. Jetson EX1 specifications: NVIDIA Jetson Xavier NX module CPU – 6-core NVIDIA Carmel ARMv8.2 64-bit processor with 6 MB L2 + 4 MB L3 cache GPU – NVIDIA Volta architecture with 384 NVIDIA CUDA cores and 48 […]

Portwell COM Express Tiger Lake-H module supports 2.5GbE, PCIe Gen 4, 8K, up to 64GB DDR4

Portwell PCOM-B657VGL COM Express TIger Lake-H

Portwell PCOM-B657VGL joins other COM Express and COM HPC modules based on Intel Tiger Lake-H Xeon, Core, and Celeron embeddded processors such as ADLINK Express-TL and Congatec conga-HPC/cTLH. Just like its competitors, the Portwell COM Express Type 6 Basic module offers 8K video output, PCIe x16 Gen 4, up to 64GB DDR4, USB 3.2 Gen 2, and 2.5GbE networking for a wide range of higher-end embedded applications such as industrial automation, medical equipment, graphics-intensive applications, and artificial intelligence. Portwell PCOM-B657VGL specifications: SoC (one or the other) General Embedded Intel Celeron 6600HE dual-core processor @ 2.6GHz with 8MB L3 Cache, 16EU Intel UHD graphics, 35W TDP Intel Core i3-11100HE processor with 4 cores / 8 threads @ 2.4GHz (up to 4.4GHz in Turbo Boost) with 8MB L3 Cache, 16EU Intel UHD graphics, 45/35W cTDP Intel Core i5-11500HE processor with 6 cores / 12 threads @ 2.6GHz (up to 4.1GHz in Turbo […]

Raspberry Pi inspired Intel SBC supports Myriad X AI accelerator, 5G connectivity

Axiomtek KIWI310

Axiomtek KIWI310 may look like a Raspberry Pi SBC but it packs an Intel Celeron N3350 processor, an M.2 slot with support for Myriad X AI accelerator, and the company also offer a HAT with 5G cellular connectivity. The single board computer also comes with up to 4GB LPDDR4 memory, up to 64GB eMMC flash, a Micro HDMI port, two USB 3.0 and two USB 2.0 ports, as well as the ubiquitous 40-pin GPIO header. Power options are also an improvement over your typical Raspberry Pi board with USB-C PD as well as LiPo battery support. Axiomtek KIWI310 specifications: SoC – Intel Celeron N3350 dual-core Apollo Lake processor @ 1.1 GHz / 2.4 GHz (Turbo) with Intel HD Graphics 500; 6W TDP System Memory – Up to 4GB (specs) or 8GB (YouTube video) LPDDR4 Storage – Up to 64GB eMMC flash, SPI flash for AMI BIOS Video & audio output […]

NXP i.MX 93 processor combines Cortex-A55 cores with Ethos U65 microNPU

NXP i.MX 93 (935x/933x)

NXP has unveiled the i.MX 93 processor family comprised of i.MX 935x, 933x, 932x, and 931x parts at this time with up to two Cortex-A55 cores, one Arm Cortex-M33 real-time core, as well as an Ethos U65 microNPU for machine learning (ML). We wrote about i.MX 9 family back in March with NXP telling us it would include an Arm Ethos U-65 microNPU and EdgeLock secure enclave, be manufactured with a 16/12nm FinFET class process, and includes the “Energy Flex” architecture to optimize power consumption by turning on/off specific blocks in the processor. The NXP i.MX 93 is the first family leveraging those new features, and we know have some more details. NXP i.MX 93 processor specifications: CPU 1x or 2x Arm Cortex-A55 @ 1.7 GHz with 32KB I-cache, 32KB D-cache, 64KB L2 cache, 256KB L3 cache with ECC 1x Arm Cortex-M33 @ 250 MHz low power microcontroller with 256KB […]

NVIDIA Jetson AGX Orin 12-core Cortex-A78E module delivers up to 200 TOPS

Jetson AGX Orin

The upcoming NVIDIA Jetson AGX Orin module packs some serious processing power with a 12-core Cortex-A78AE processor, 2048 CUDA cores, and 64 Tensor cores delivering up to 200 TOPS of AI performance, or 6 times more than its predecessor, the Jetson AGX Xavier module. Designed for robotics, autonomous machines, medical devices, the Jetson AGX Orin delivers the same performance as a GPU-enabled server, but in a much more compact 100 x 87mm form factor. The module can be operated in three power modes, namely 15W, 30W, or 50W depending on performance and power requirements, and a compact developer kit will also be made available. Jetson AGX Orin specifications: CPU – 12-core Arm Cortex-A78AE v8.2 64-bit processor with 3MB L2 + 6MB L3 cache GPU / AI accelerators NVIDIA Ampere architecture with 2048 NVIDIA CUDA cores and 64 Tensor Cores @ 1 GHz DL Accelerator – 2x NVDLA v2.0 Vision Accelerator […]