STMicro STM32MP23 Cortex-A35/M33 MPU features a 600 GOPS NPU for industrial IoT and ML applications

STM32MP235 block diagram

STMicroelectronics STM32MP23 is a dual-core Cortex-A35 general-purpose microprocessor (MPU) with a Cortex-M33 real-time core, and 600 GOPS of performance operating at up to 125°C and designed for industrial and Internet-of-Things (IoT) edge computing, advanced HMI, and machine-learning (ML) applications. It follows the STM32MP25 series unveiled in 2023 and launched the following year as the first member of the STM32MP2 family. The STM32MP23 is a cost-down version of the STM32MP25 with similar specifications (and pin-to-pin compatible packages), but the new family has a weaker 0.6 TOPS NPU, and tops at two Ethernet ports for the high-end parts instead of three. Ditto for CAN Bus interfaces (2x vs 3x). Other changes include a 16-bit memory interface and the lack of PCIe and USB 3.0 interfaces. STM32MP23 specifications: CPU – Single or dual Arm Cortex-A35 cores running at up to 1.5 GHz or 1.2 GHz with 512KB L2 cache Real-time MCU – Arm […]

Radxa Orion O6 Preview – Part 2: Debian 12 – What works, what doesn’t

Orion O6 Review Debian 12

I went through an unboxing and Debian 12 installation on the Radxa Orion O6 at the end of January, but decided to work on other reviews since software support still needed to be worked on. Since then, there’s been some work done, but no new image released. After waiting for almost two months, I’ve decided to carry on with the review by testing the Debian 12 image in a way similar to the Rock 5B SBC preview I did with Debian 11 in 2022 to check what works and what doesn’t on the Orion O6 at the time of the review. That will involve testing all ports, including 5GbE networking and the PCIe slot with an (old) NVIDIA graphics card, running some benchmarks, and also trying the Debian 12 image with a self-built Linux 6.13 kernel using ACPI instead of UEFI for the default image. Orion O6 SBC benchmarks on […]

Imagination DXTP GPU promises 20% higher power efficiency compared to DXT GPU for mobile devices and laptops

Imagination DXTP power-efficient GPU

Imagination has announced the DXTP GPU for mobile and power-constrained devices with up to 20% greater power efficiency than its predecessor, the D-Series DXT GPU for longer battery life for AI applications or complex games. The company says the GPU is suitable for gaming, user interfaces, graphics-rich applications, computer vision, generative AI, and other AI applications designed to run on smartphones, tablets, laptops, desktops, or non-safety automotive products. Imagination DXTP highlights: Two off-the-shelf configurations: DXTP-48-1536 48 GPixel/s 1.5 TFLOPS FP32 (1536 FLOPs/Clock) 3 TFLOPS FP16 6 TOPS INT8 48 ppc Bilinear 3D Textured Up to 96 ppc 2D Dual Rate Mode DXTP-64-2048 64 GPixel/s 2 TFLOPS FP32 (2048 FLOPs/Clock) 4 TFLOPS FP16 8 TOPS INT8 64 ppc Bilinear 3D Textured Up to 128 ppc 2D Dual Rate Mode Bus Interface – AXI, ACE-Lite Compression PVRIC5 Lossless and Lossy Framebuffer Compression ETC and ASTC LDR and HDR Support Microarchitecture improvement – […]

Phison’s aiDAPTIV+ AI solution leverages SSDs to expand GPU memory for LLM training

Phison's aiDAPTIVCache family support 70B model

While looking for new and interesting products I found ADLINK’s DLAP Supreme series, a series of Edge AI devices built around the NVIDIA Jetson AGX Orin platform. But that was not the interesting part, what got my attention was it has support for something called the aiDAPTIV+ technology which made us curious. Upon looking we found that the aiDAPTIV+ AI solution is a hybrid (software and hardware) solution that uses readily available low-cost NAND flash storage to enhance the capabilities of GPUs to streamline and scale large-language model (LLM) training for small and medium-sized businesses. This design allows organizations to train their data models on standard, off-the-shelf hardware, overcoming limitations with more complex models like Llama-2 7B. The solution supports up to 70B model parameters with low latency and high-endurance storage (100 DWPD) using SLC NAND. It is designed to easily integrate with existing AI applications without requiring hardware changes, […]

Disabling VT-d improves Intel Arc GPU Linux performance on Meteor Lake and newer SoCs

Improve Intel Arc GPU performance Linux VT-d disabled

In this post, I’ll check whether disabling VT-d virtualization support may improve the performance of the Intel Arc GPU in recent Meteor Lake or Lunar Lake SoC using a Khadas Mind Maker Kit with an Intel Core Ultra 7 258V CPU with Intel Arc 140V graphics running Ubuntu 24.10. A few days ago, I read a post on Phoronix about Intel publishing tips to improve the performance of Intel GPUs in Linux: Keep the system updated with the latest kernel and Mesa versions. Ensure SoC firmware is up-to-date. These firmware updates currently require installing the Windows graphics driver; firmware updates via fwupd are in progress. Use Wayland where possible, as it supports additional modifiers for better performance. For MTL (Meteor Lake) and newer integrated GPUs, disable VT-d if virtualization is not needed. For discrete GPUs: Enable ReBAR_ Enable ASPM_ I was especially curious about the line about disabling VT-d virtualization […]

Khadas Mind Maker Kit review – Part 2: Windows 11 Home on an Intel Core Ultra 7 258V AI mini PC

Khadas Mind Maker Kit Review Windows 11

I’ve already gone through the specifications and an unboxing of the “Khadas Mind 2 AI Maker Kit” powered by an Intel Core Ultra 7 258V “Lunar Lake” processor delivering up to 115 TOPS of AI performance and equipped with 32GB LPDDR5X RAM and a 1TB M.2 NVMe SSD in the first part of the review. I’ve now spent time with the mini PC/developer kit which is now simply called “Khadas Mind Maker Kit”, and I will report my experience with the Windows 11 Home 24H2 operating system in the second part of the review testing features, running benchmarks including an AI benchmark, evaluating networking and storage performance, testing the thermal design while under stress, and taking measurements for fan noise and power consumption. It looks like some AI features may finally be usable on Windows, but I’ll test that in a separate post since everything is new and Microsoft Copilot+, […]

Vulkan 1.4 3D graphics and compute API released

Vulkan 1.4

The Khronos Group has just announced the release of Vulkan 1.4 cross-platform 3D graphics and compute API. The new release makes some of the optional extensions and features mandatory, adds streaming transfers, and supports 8K rendering on up to eight targets. Minimum hardware limits have also been increased including at least seven maxBoundDescriptorSets and eight maxColorAttachments. Vulkan 1.4 highlights: Streaming Transfers: new implementation requirements to ensure applications can stream large quantities of data to a device while simultaneously rendering at full performance. Previously optional extensions and features critical to emerging high-performance applications are now mandatory in Vulkan 1.4, ensuring availability across multiple platforms. These include push descriptors, dynamic rendering local reads, and scalar block layouts. Maintenance extensions up to and including VK_KHR_maintenance6 are now part of the core Vulkan 1.4 specification. 8K rendering with up to eight separate render targets is now guaranteed to be supported, along with several other […]

DietPi 9.7 and Armbian 24.8 released with improved support for Rockchip, Amlogic, and Allwinner SBCs

DietPi 9.7 Orange Pi 5 Plus

Armbian and DietPi are two separate projects that provide Linux-based OS images optimized for Arm-based single board computers. The last time we had a look at both projects was in June with the release of Armbian 24.5.1 and DietPi 9.4, but there have been several updates since then including the releases of the latest DietPi 9.7 and Armbian 24.8 Yelt just a few days ago. So let’s check out the latest changes. DietPi 9.7 DietPi is a lightweight Debian-based Linux distribution for SBCs and server systems that ships as a minimal image but users can install any packages they want, including the ones required for desktop environment, to match the requirements of the applications. It’s notably used by the Linamp project – a Raspberry Pi 4-based project that brings WinAMP to real life – that we covered a few weeks ago. DietPi 9.7 was released on August 25, 2024 with […]

UP 7000 x86 SBC