artificial intelligence News - Page 15 of 100 - CNX Software

August 30, 2023November 11, 2023 by Jean-Luc Aufranc (CNXSoft) - 8 Comments

Ugoos AM8 – A true 8K TV box powered by Amlogic S928X-J processor

Ugoos AM8 is a true 8K TV box based on Amlogic S928X-J penta-core Cortex-A76/A55 processor with Dolby Vision and Dolby Audio support that is now available for sale on Aliexpress. If you search for an 8K TV box on the web, you’ll find plenty of listings for 4K TV boxes incapable of 8K video output, but Ugoos AM8 is what I would call “a true 8K TV box” capable of both video playback and output up to 8Kp60 thanks to the Amlogic S928X processor which we previously found announced in IPTV/OTT devices for operators from SDMC and SEI Robotics, as well as more recently in an 8K TV box board. Ugoos AM8 (preliminary) specifications: SoC – Amlogic S928X-J CPU – Penta-core processor with 1x Cortex-A76 core, 4x Cortex-A55 cores GPU – Arm Mali-G57 MC2 GPU with support for OpenGL ES 3.2, Vulkan 1.2, and OpenCL 2.0 VPU 8Kp60 H.265, AV1, […]

August 25, 2023 by Jean-Luc Aufranc (CNXSoft) - 6 Comments

Modular Mojo claims to be over 36,000 times faster than Python for AI workloads

Modular Mojo is a new programming language designed for AI developers that is said to combine the usability of Python with the performance of C with over 36,000 times the performance of Python on a matrix multiplication workload. Modular Mojo programming language was not in the initial plan of the company but came about when the company’s founders – who focused on building a platform to unify the world’s ML/AI infrastructure – realized that programming across the entire stack was too complicated and also ended up writing a lot of MLIR (Multi-Level Intermediate Representation) by hand. The “over 36,000 times speedup” claim comes with the matmul.py script performing a 128×128 matrix multiplication in Python with a throughput of 0.00215 GFLOP/s and another script doing 512×512 vectorized + parallelized matrix multiplication in Mojo at 79.636 GFLOP/s. The claim looks dubious and that’s odd they used different matrix sizes, but some are […]

August 25, 2023January 9, 2024 by Jean-Luc Aufranc (CNXSoft) - 1 Comment

Youyeetoo YY3568 devkit review – Part 1: Unboxing, specifications, and Android 11 testing

Youyeetoo has sent us a review sample of their YY3568 “Bundle 5” devkit with the Rockchip RK3568-powered YY3568 SBC, an 11.6-inch touchscreen display, a MIPI camera module, and all accessories required to get started. We were especially interested in using it to play with the 1 TOPS NPU in the Rockchip RK3568 in Linux, but we’ll start the Youyeetoo YY3568 review with an unboxing, some specifications, and a quick review with Android 11 before switching to Debian 10 in the second part of the review. Youyeetoo YY3568 devkit unboxing The YY3568 single board computer itself is comprised of a carrier board and a YY3568-Core board with a Rockchip RK3568 processor, as well as 8GB RAM, 64GB eMMC flash, and WiFi 5 and Bluetooth 5.0 module. The board is suitable for various applications from generic computing to video playback and AI workloads. YY3568 SBC specifications and hardware overview Youyeetoo YY3568 board […]

August 22, 2023August 22, 2023 by Jean-Luc Aufranc (CNXSoft) - 2 Comments

Tiny solder-down NXP i.MX 93 System-on-Module powers credit card-sized evaluation board

Ka-Ro Electronics’ QS93 is a tiny solder-down NXP i.MX 93 System-on-Module (SoM) running Linux and designed for edge processing. The company also offers a credit card-sized evaluation board that may remind some of the Raspberry Pi with its GPIO header and general layout, but it comes with two Fast Ethernet ports and one USB 2.0 port. We’ve already covered several system-on-modules based on the NXP i.MX 93 Cortex-A55/M33 AI processor including some with high-density board-to-board connectors such as the Compulab UCM-IMX93 and Forlinx FET-MX9352-C, others with a SO-DIMM connector like the VAR-SOM-MX93, and finally some designed to be soldered on the carrier board such as the OSM-L compatible iW-RainboW-G50M, and the QS93 adds to the latter category in a tiny 27×27 mm form factor. Ka-Ro electronics QS93 specifications: SoC – NXP i.MX 93 with CPU – Up to dual-core Cortex-A55 processor @ up to 1.5 GHz Real-time core – Arm […]

August 21, 2023August 21, 2023 by Jean-Luc Aufranc (CNXSoft) - 1 Comment

MediaPipe for Raspberry Pi released – No-code/low-code on-device machine learning solutions

Google has just released MediaPipe Solutions for no-code/low-code on-device machine learning for the Raspberry Pi (and an iOS SDK) following the official release in May for Android, web, and Python, but it’s been years in the making as we first wrote about the MediaPipe project back in December 2019. The Raspberry Pi port is an update to the Python SDK and supports audio classification, face landmark detection, object detection, and various natural language processing tasks. MediaPipe Solutions consists of three components: MediaPipe Tasks (low-code) to create and deploy custom end-to-end ML solution pipelines using cross-platform APIs and libraries MediaPipe Model Maker (low-code) to create custom ML models MediaPipe Studio (no-code) webpage to create, evaluate, debug, benchmark, prototype, and deploy production-level solutions. You can try it out directly in your web browser at least on PC and I could quickly test the object detection on Ubuntu 22.04. MediaPipe Tasks can be […]

August 7, 2023August 10, 2023 by Sponsored Post - No Comments

Get Early Access to the World’s Most Powerful Edge AI Accelerator! (Sponsored)

AxeleraAI EAP Evaluation Kit Box and System

Final Call for Axelera AI’s Early Access Program A few months ago Axelera AI launched their Metis AI hardware and software platform, available as M.2 cards, PCIe boards, and vision-ready systems, combined with the Voyager SDK. Now, Axelera AI’s Early Access Program (EAP) is entering its final phase – the first evaluation kits are almost ready to be shipped. This program offers a unique opportunity for customers who have signed up to experience firsthand the accuracy, performance, and usability of the Metis AI Platform. Axelera AI delivers the most powerful AI Accelerator in the world. With a PCIe AI Edge accelerator card delivering a peak performance of 214 TOPS. Using a smaller form factor, the M.2 AI Edge accelerator module delivers up to 106 TOPS peaks. This handles the most demanding vision applications. In this article, we will delve into the advantages of the EAP by sharing three customer stories […]

August 4, 2023 by Jean-Luc Aufranc (CNXSoft) - No Comments

Hailo-8L 13 TOPS AI accelerator targets entry-level edge devices

Hailo introduced the Hailo-8 AI accelerator offering up to 26 TOPS in 2020, and we’ve found it integrated into many designs since then. The company has now launched a cost-down version with the Hailo-8L AI accelerator delivering up to 13 TOPS for more cost-sensitive entry-level edge devices, or workloads that do not require the more powerful Hailo-8. Hailo says the Hailo-8L offers low-latency, high-efficiency processing, and is capable of handling pipelines with multiple real-time streams and concurrent processing of multiple models and AI tasks. The new Hailo-8L is compatible with the Hailo-8 and relies on the same Hailo-8 software suite, so they could be integrated into existing designs for cost savings. Hailo-8L key features and specifications: 13 Tera-Operations Per Second (TOPS) Real-time, low latency & high-efficiency AI inferencing on edge devices No external memory required Scalable with simultaneous processing of multi-streams & multi-models Typical power consumption – 1.5W Commercial & […]

July 28, 2023August 2, 2023 by Jean-Luc Aufranc (CNXSoft) - No Comments

Tungsten700 SMARC SoM and devkit features MediaTek Genio 700 AIoT processor

Laird Connectivity Tungsten700 SOM is a SMARC system-on-module powered by a MediaTek Genio 700 Arm Cortex-A78/A55 AIoT processor with up to 8GB LPDDR4, 16GB eMMC flash, and a Sona MT320 Wi-Fi 6/Bluetooth 5.3 module based on the Filogic 320 chipset. The board was designed by Boundary Devices, recently acquired by Laird, and is offered with a SMARC 2.1 carrier board that can be used for development or as a single board computer integrated into designs. Tungsten700 SMARC module Tungsten700 specifications: SoC – MediaTek Genio 700 (MT8390) CPU – Octa-core processor with 2x Arm Cortex-A78 cores @ up to 2.2 GHz, 6x Arm Cortex-A55 cores @ up to 2.0 GHz GPU – ARM Mali-G57 MC3 GPU VPU as in “Video Processing Unit” Encode up to 4Kp30 HEVC/H.264 Decode up to 4Kp75 HEVC/H.264/AV1/VP9 VPU as in “Vision Processing Unit” – Tensilica VP6 Vision Processing Unit ISP Single Camera: 32MP @ 30FPS Dual […]