Posted on by Jean-Luc Aufranc (CNXSoft) - No Comments on Using the Khadas Mind Graphics 2 (NVIDIA RTX 5060 Ti USB4 eGPU) on Ubuntu 26.04

Using the Khadas Mind Graphics 2 (NVIDIA RTX 5060 Ti USB4 eGPU) on Ubuntu 26.04

Khadas Mind Graphics 2 NVIDIA GeForce RTX 5060 Ti 16GB eGPU Ubuntu 26.04

When we used the NVIDIA GeForce RTX 5060 Ti 16GB-based Khadas Mind Graphics 2 dock with a mini PC running Ubuntu 26.04 last week, the USB4-connected eGPU was detected and enabled automatically, but we encountered two issues: very poor performance and the mini PC failed to boot when the dock was removed. The solution is partially documented as part of a long GEEKOM IT13 Max Ubuntu 26.04 review, but I thought it might be useful to document everything in a shorter article for people who own the Khadas Mind Graphics 2 or another NVIDIA Blackwell eGPU connected over USB4/Thunderbolt. Fixing Khadas Mind Graphics 2 performance issues on Ubuntu The Khadas Mind Graphics 2 and NVIDIA graphics card were automatically detected when I connected the dock to an Ubuntu 26.04 mini PC

Posted on by Jean-Luc Aufranc (CNXSoft) - No Comments on GEEKOM IT13 Max Review – Part 3: Ubuntu 26.04 on an Intel Core Ultra 9 185H mini computer

GEEKOM IT13 Max Review – Part 3: Ubuntu 26.04 on an Intel Core Ultra 9 185H mini computer

GEEKOM IT13 Max Mini PC review Ubuntu 26.04 Linux

We’ve already checked out the hardware with an unboxing and a teardown of the GEEKOM IT13 Max in the first part of the review, before testing the Intel Core Ultra 9 185H mini PC with Windows 11, so it’s now time to test Linux on the mini computer using Ubuntu 26.04. In this review, we will report our experience running Ubuntu 26.04 on the GEEKOM IT13 Max, after going through feature testing, running system benchmarks and AI benchmarks on the CPU, GPU, and NPU, checking out storage and USB performance, testing 2.5GbE and WiFi 6/7, stress testing the mini PC, and measuring the Meteor Lake mini PC’s fan noise and power consumption. Installing Ubuntu 26.04 We installed Ubuntu 26.04 alongside Windows 11 in dual boot configuration. We first shrank the Windows partition (C:) size by approximately half, before inserting a USB flash drive with Ubuntu 26.04 to start the installation […]

Sponsored
Sponsored
Sponsored
Posted on by Debashis Das - 6 Comments on EITWatch open-source ESP32-S3 smartwatch implements planar EIT hand gesture recognition

EITWatch open-source ESP32-S3 smartwatch implements planar EIT hand gesture recognition

EITWatch

Researchers from Northwestern University have introduced EITWatch, an open-source ESP32-S3-based EIT (Electrical Impedance Tomography) gesture-sensing system that fits within a standard smartwatch footprint. It relies on a Seeed Studio XIAO ESP32-S3 board and eight electrodes arranged on the back of the case to detect hand gestures via impedance changes caused by muscle and tendon movement beneath the skin. Unlike other wrist-EIT systems that use circumferential electrode bands, EITWatch places all eight electrodes on the back of the smartwatch case. The sensing area consists of a 31 mm electrode ring that fits within a standard 40 mm watch case, while a 20 mm PCB extension carries the discrete analog front end (AFE). EITWatch hardware specifications: Wireless board – XIAO ESP32-S3 SoC – Espressif ESP32-S3R8 CPU – Dual-core Tensilica LX7 microcontroller up to 240 MHz with vector instructions for AI acceleration Memory – 512KB SRAM, 8MB PSRAM Wireless – Wi-Fi 4 and […]

Posted on by Jean-Luc Aufranc (CNXSoft) - 2 Comments on Giada DK320 digital signage player and Edge AI embedded PC supports up to the Intel Core 5 315 Wildcat Lake processor

Giada DK320 digital signage player and Edge AI embedded PC supports up to the Intel Core 5 315 Wildcat Lake processor

Giada DK320

Giada DK320 is an ultra-compact Wildcat Lake mini PC powered by up to an Intel Core 5 315 hexa-core processor and designed for digital signage, edge computing, and embedded applications requiring real-time analytics, intelligent monitoring, and automated decision-making. The system supports up to 48GB DDR5 RAM and M.2 2240/2280 NVMe storage, and offers dual 4K-capable HDMI video output, 2.5GbE networking, an optional WiFi 5/6/7 and Bluetooth module, an RS232 COM port, and four USB 3.2 ports. Giada DK320 specifications: Wildcat Lake SoC (one or the other) Intel Core 3 304 5-core CPU – 1x P-cores @ 1.5/4.3 GHz (Turbo) + 4x LPE-cores @ 1.4/3.3 GHz (Turbo) GPU – 1-core Intel Xe3 Graphics @ 2.3 GHz (9 TOPS) NPU – 15 TOPS Intel Core 5 315 6-core CPU – 2x P-cores @ 1.5/4.4 GHz (Turbo) + 4x LPE-cores @ 1.4/3.3 GHz (Turbo) GPU – 2-core Intel Xe3 Graphics @ 2.3 GHz […]

Posted on by Jean-Luc Aufranc (CNXSoft) - 3 Comments on Gemma Translator multilingual interpreter runs locally on Raspberry Pi 5 with the LiteRT runtime

Gemma Translator multilingual interpreter runs locally on Raspberry Pi 5 with the LiteRT runtime

Gemma Translator

Gemma Translator is an open-source project for the Raspberry Pi 5 that acts as a multilingual interpreter running locally without access to the cloud using Google Gemma 4 and LiteRT runtime. It’s a project built by three Google engineers with the help of Google Antigravity. It includes code to run an on-device, fully offline voice translator/interpreter using Gemma 4 E2B (2.3B effective / ~5.1B total parameters with per-layer embeddings) and the LiteRT runtime. The hardware is comprised of a few components: A Raspberry Pi 5 with 8GB of RAM Display – HDMI monitor or touchscreen display such as a 480×320 kiosk display Audio Microphone or USB audio capture interface Speaker or headphone output device Keyboard  – The project demonstrated below appears to use a custom PCB (AX-4LABS-NEW) with a rotary encoder and four mechanical keys (of which only two are used) Optional 3D printed parts for the enclosure Software highlights: […]

Posted on by Jean-Luc Aufranc (CNXSoft) - 1 Comment on AAEON PICO-ADN2 – A low-profile Alder Lake-N Pico-ITX SBC for space-constrained applications

AAEON PICO-ADN2 – A low-profile Alder Lake-N Pico-ITX SBC for space-constrained applications

AAEON PICO ADN2

AAEON PICO-ADN2 is a low-profile Alder Lake-N Pico-ITX SBC powered by an Intel N97 or Core i3-N305 SoC and designed for space-constrained transportation, digital signage, industrial automation, and smart kiosk applications. Its features are very similar to the PICO-ADN4 Pico-ITX SBC introduced in 2023, with up to 16 GB LPDDR5, HDMI, LVDS, and eDP display interfaces, dual Gigabit Ethernet, an M.2  E-Key socket for wireless, multiple USB interfaces, RS232/RS485 interfaces, and more. The main difference is that most standard (and thick) ports have been replaced by low-profile connectors, except somehow the HDMI connector. AAEON PICO-ADN2 specifications Alder Lake-N SoC (one or the other) Intel Processor N97 quad-core processor up to 3.6 GHz with 6MB cache, 24EU Intel UHD Graphics; TDP: 12W Intel Core i3-N305 octa-core processor up to 3.8 GHz with 6MB cache, 32EU Intel UHD Graphics; TDP: 15W Other Intel Atom x7000E Series, Processor N-series, and Intel Core i3 […]

Advertisement

Posted on by Jean-Luc Aufranc (CNXSoft) - 17 Comments on Qualcomm Snapdragon C Arm SoC outperforms an Intel N250 CPU by up to 67 percent in “unplugged” benchmarks

Qualcomm Snapdragon C Arm SoC outperforms an Intel N250 CPU by up to 67 percent in “unplugged” benchmarks

Qualcomm Snapdragon C specs benchmarks

Qualcomm unveiled the Snapdragon C entry-level Arm SoC for $300+ Windows laptops at Computex 2026 with a ridiculously small amount of details last May, so we had to rely on the specs of the upcoming Acer Aspire Go 15 (AG15-Q31P) laptop. Qualcomm has now released a document entitled “Snapdragon C Platform Overview” with more technical details and “unplugged” benchmarks – meaning running on batteries – comparing a Snapdragon C laptop to one based on Intel Processor N250 Twin Lake SoC. The presentation Snapdragon C specifications: CPU – Up to an octa-core Kryo processor at up to 3.0 GHz (single-core) or 2.0 GHz (quad-core) with 2 MB total cache; note: no data for eight cores GPU – Qualcomm Adreno GPU at up to 900 MHz AI accelerator – Built-in Qualcomm Hexagon NPU (still unspecified performance) System Memory – Up to 16GB LPDDR5/5x or LPDDR4 Storage – PCIe 3.0 NVMe, UFS 2.2/3.1 […]

Posted on by Jean-Luc Aufranc (CNXSoft) - No Comments on WCH CH32V407/467 RISC-V MCU integrates Fast Ethernet MAC + PHY, 480 Mbps USB 2.0 PHY, up to 8 MB on-chip PSRAM

WCH CH32V407/467 RISC-V MCU integrates Fast Ethernet MAC + PHY, 480 Mbps USB 2.0 PHY, up to 8 MB on-chip PSRAM

CH32V407 block diagram

WCH CH32V407 and CH32V467 are new 200 MHz RISC-V microcontrollers with a built-in 10/100 Mbps Ethernet MAC + PHY, a USB 2.0 high-speed (480 Mbps) host PHY, up to 8 MB on-chip PSRAM, and a range of peripherals for industrial IoT & edge AI applications. Both are based on the company’s Qingke V3V core with RVV vector extensions, delivering 4.11 CoreMark/MHz, a score higher than a typical Arm Cortex-M4 core. The main difference between the CH32V407 and CH32V467 is that the latter comes with 4 MB or 8 MB PSRAM, making it more suitable for display/HMI applications. WCH CH32V407/467 specifications: MCU Core – QingKe V3V 32-bit RISC-V (IMACV-X, X = custom) core @ up to 200 MHz RVV-vector instruction set and parallel computing Memory and storage 200KB SRAM 4MB or 8MB PSRAM memory (CH32V467 only) 512KB Flash SDIO Interface (MMC, SD/SDIO Cards, and CE-ATA) Optional Flexible Static Memory Controller (FSMC) […]

Boardcon MINI1126B-P AI vision system-on-module wit Rockchip RV1126B-P SoC