Review of GEEKOM A5 Pro 2026 Edition (AMD Ryzen 5 7530U) mini PC – Part 1: specifications, unboxing, and teardown

We’ve just received a sample of the GEEKOM A5 Pro 2026 Edition mid-range mini PC for review. It’s powered by an AMD Ryzen 5 7530U 6-core/12-thread processor clocked at up to 4.5GHz, and equipped with 16GB dual-channel DDR4-3200MHz SODIMM memory and a 1TB NVMe SSD. The Windows 11 Pro computer features two 4K-capable HDMI 2.0 ports, two USB-C ports with DisplayPort Alt mode, a 2.5GbE RJ45 jack, a Wi-Fi 6 and Bluetooth 5.2 module, a full-size SD card reader, and a few USB 32/2.0 Type-A ports. In the first part of the review, we’ll go through the specifications, an unboxing,...

TerraMaster F2-425 Plus 3+2-bay hybrid NAS review – Part 1: unboxing, teardown, drives installation, and first boot

TerraMaster has sent us a sample of the F2-425 Plus 3+2-bay NAS for review. It’s powered by an Intel Processor N150 CPU paired with 8GB of RAM, with two 3.5-inch SATA bays and three M.2 NVMe sockets for storage, as well as two 5GbE RJ45 jacks, and a few USB ports. Since we’ve already written the specifications in the previous article, we’ll focus the first part of the review on an unboxing and a teardown to check out the hardware design, before showing how to install the drives, and giving it a quick first try. TerraMaster F2-425 Plus NAS unboxing...

Frigate with Hailo for object detection on a Raspberry Pi

I run Frigate to record security cameras and detect people, cars, and animals when in view. My current Frigate server runs on a Raspberry Pi CM4 and a Coral TPU plugged in via USB.

Raspberry Pi offers multiple AI HAT+'s for the Raspberry Pi 5 with built-in Hailo-8 or Hailo-8L AI coprocessors, and they're useful for low-power inference (like for image object detection) on the Pi. Hailo coprocessors can be used with other SBCs and computers too, if you buy an M.2 version.

OpenTitan’s Transition to Side-Channel Resilient Post-Quantum Cryptography

...

AsteroidOS 2.0 open-source smartwatch OS released, now supports around 30 devices

AsteroidOS 2.0 Linux-based, open-source smartwatch operating system has just been released with features such as always-on display support, Tilt-to-Wake, a customizable QuickPanel, multiple launcher styles, Nightstand mode, performance improvements, and support for about 30 devices. It’s been a long journey. We first noted the open-source project in 2016 when Florent Revest showcased a basic user experience on the LG G Watch, just before giving a talk at FOSDEM 2016 introducing AsteroidOS. This was followed by the launch of the Connect Watch (crowdfunded) AsteroidOS smartwatch in 2017 by a French company (which didn’t get funded), and AsteroidOS 1.0 was released in 2018....

Murena Volla 12.6-inch privacy-focused tablet runs /e/OS Google-free Android OS on MediaTek Helio G99 SoC

Murena Volla is a 12.6-inch privacy-focused tablet running /e/OS and designed for users seeking a Google-free experience. While the hardware is developed in collaboration with German-based Volla, the Murena edition comes preloaded with the Android-based /e/OS and integrated Murena cloud services for users who want full control over their personal data. The tablet is built around a MediaTek Helio G99 octa-core processor paired with 12GB of RAM and 512GB of internal storage. The device also features a USB Type-C port, Wi-Fi 6, 4G LTE support (Europe only), and 13MP rear and 5MP front-facing cameras. Applications include secure business use, education,...

NXP S32N79 octa-core Arm Cortex-A78E/12-core Cortex-R52 “Super-Integration Processor” targets Software-Defined Vehicles (SDV)

NXP recently introduced the S32N79 “Super-Integration” automotive processor, part of the S32N7 series, equipped with up to eight Arm Cortex-A78E application cores and twelve Arm Cortex-R52 cores for real-time processing. Building on the earlier 5 nm S32N55 16-core Cortex-R52 + 2x Lockstep Cortex-M7 automotive processor, the S32N79 automotive processor is still designed for software-defined vehicles (SDV), but its Cortex-A78E applications cores further enable features such as ADAS sensor fusion and data AI services, as well as improved vehicle gateway/processing functions. NXP S32N79 key features and specifications: CPU Up to 8x split-lock Arm Cortex-A78AE cores operating at up to 1.8 GHz...

ESP32-P4-PC Open Source Hardware Board – the most comprehensive and feature-rich ESP32-P4 board on the market

Introducing the ESP32-P4-PC Open Source Hardware Board – the most comprehensive and feature-rich ESP32-P4 board on the market. Built around the powerful dual-core ESP32-P4 RISC-V SoC, this board delivers serious performance and unmatched peripheral integration for advanced embedded and IoT development. From HMI systems and industrial controllers to edge AI and vision applications — the […]

AI is destroying Open Source, and it's not even good yet

Over the weekend Ars Technica retracted an article because the AI a writer used hallucinated quotes from an open source library maintainer.

The irony here is the maintainer in question, Scott Shambaugh, was harassed by someone's AI agent over not merging its AI slop code.

It's likely the bot was running through someone's local 'agentic AI' instance (likely using OpenClaw). The guy who built OpenClaw was just hired by OpenAI to "work on bringing agents to everyone." You'll have to forgive me if I'm not enthusastic about that.

Instruction decoding in the Intel 8087 floating-point chip

In the 1980s, if you wanted your IBM PC to run faster, you could buy the Intel 8087 floating-point coprocessor chip. With this chip, CAD software, spreadsheets, flight simulators, and other programs were much speedier. The 8087 chip could add, subtract, multiply, and divide, of course, but it could also compute transcendental functions such as tangent and logarithms, as well as provide constants such as π. In total, the 8087 added 62 new instructions to the computer.

But how does a PC decide if an instruction was a floating-point instruction for the 8087 or a regular instruction for the 8086 or 8088 CPU? And how does the 8087 chip interpret instructions...