SpacemiT K3 Pico-ITX RISC-V Chassis Kit Review – Part 2: What works, what doesn’t in Bianbu OS 4.0
Last month, I received the SpacemiT K3 Pico-ITX Chassis Kit based on the company's K3 16-core RISC-V SoC, and started the review with an unboxing, a teardown, and a first boot to Bianbu OS 4.0. Since the system features a 10Gbps Ethernet SFP+ cage, I also had to order a 10GbE SFP+ to Copper adapter, as my 10GbE networking gear is exclusively based on RJ45 ports. In this review, I'll check system information in Bianbu OS 4.0.1, run a few benchmarks, test 10GbE, GbE, and WiFi 6 networking performance, play YouTube videos at various resolutions, run AI workloads (LLM), check all/most features work as expected, and measure the power consumption of the SpacemiT K3 "Pico-ITX Chassis Kit" mini PC. Bianbu OS 4.0 System Information Before running anything, I updated the system: 594 packages were updated. For reference, the update also involved updating the EC (Embedded Controller) firmware: Let's double-check the
UP WCL – A credit card-sized Wildcat Lake SBC with up to 24GB LPDDR5, 256GB UFS
AAEON's UP WCL is an upcoming credit card-sized single board computer (SBC) powered by an Intel Wildcat Lake processor up to the Core 7 350 hexa-core CPU paired with up to 24GB LPDDR5 and 256GB UFS storage. The board also features an HDMI 2.1 video output, a 2.5GbE RJ45 jack, an M.2 Key-E socket for optional WiFi and Bluetooth connectivity, three USB 3.2 ports, and a few wafers for I/O expansion. AAEON UP WCL 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 5 320 6-core CPU – 2x P-cores @ 1.5/4.6 GHz (Turbo) + 4x LPE-cores @ 1.4/3.4 GHz (Turbo) GPU – 2-core Intel Xe3 Graphics @ 2.5 GHz (20 TOPS) NPU –
Amlogic A311Y3 octa-core Edge AI SoC features Cortex-A78/A55 cores, 8 TOPS NPU, LPDDR5 support
Amlogic has introduced the A311Y3 Edge AI SoC, which seems to be an upgrade over earlier designs such as the A311D and A311D2 used in boards like the Khadas VIM3 and Khadas VIM4. Although not listed on the official Amlogic website, documentation from vendors such as Shenzhen Tomato Technology indicates that the A311Y3 SoC is built around a 6nm process node and features a newer CPU architecture based on Arm Cortex-A78/A55 cores. The SoC integrates an NPU with up to 8 TOPS of compute performance and supports LPDDR5/LPDDR5X memory, which can be used in both processing and Edge AI, compared to earlier A311-series chips. Amlogic A311Y3 specifications CPU 2x Arm Cortex-A78 cores (64KB L1 I-cache, 64KB L1 D-cache, 256KB L2 cache) 6x Arm Cortex-A55 cores (32KB L1 I-cache, 32KB L1 D-cache, 128KB L2 cache) with a shared 2MB L3 cache Neon and Crypto extensions included Co-processors – RISC-V core for system control processing; Built-in
ADLINK COM-HPC-mPTL COM-HPC Mini computer-on-module features up to 180 TOPS Intel Core Ultra Series 3 CPU
ADLINK COM-HPC-mPTL is a COM-HPC R1.3 Mini computer-on-module powered by the Intel Core Ultra Processor Series 3 "Panther Lake" family, up to the 180 TOPS 16-core Core Ultra X7 358H processor. The module supports up to 64GB LPDDR5x and optional NVMe SSD BGA storage, and features two 2.5GbE controllers, optional MIPI CSI camera connectors, and a 40-pin debug connector. All I/Os are exposed through a standard 400-pin high-density board-to-board connector, including four display interfaces (HDMI, DP, USB4, eDP), and up to 16x PCIe Gen4/Gen5 lanes. COM-HPC-mPTL specifications: SoC – Intel Core Ultra Series 3 Panther Lake-H processors (one or the other) Intel Core Ultra 5 325 8-core (4P+0E+4LPE) processor up to 2.1 GHz / 4.5 GHz (Turbo) with 12MB cache, 4-core Xe3 graphics (40 TOPS), 47 TOPS NPU; PBP: 25W Intel Core Ultra 7 356H 16-core (4P+8E+4LPE) processor up to 1.9 GHz / 4.7 GHz (Turbo) with 18MB cache, 4-core
$4,000 AMD Ryzen AI Halo Developer Platform features 126 TOPS Ryzen AI Max+ 395 processor
Last month, AMD introduced the Ryzen AI Halo Developer Platform based on the Ryzen AI Max+ 395 processor, delivering up to 126 TOPS, and pre-orders are now available for $3,999.99 at MicroCenter (US only, in-store pickup). The computer ships with 128GB LPDDR5x memory and a 2TB NVMe SSD. It offers 10GbE and WiFi 7 networking, HDMI 2.1 and USB-C DisplayPort video output, and a few extra USB-C ports Ryzen AI Halo developer kit specifications: SoC - AMD Ryzen AI Max+ 395 CPU - 16 cores, 32 threads, “Zen 5” Architecture clocked up to 5.1 GHz GPU - AMD Radeon 8060S Integrated Graphics with 40 CUs, RDNA 3.5 architecture NPU - AMD XDNA™ 2 NPU TDP - 120W Process - TSMC 4nm FinFET Memory - 128GB LPDDR5x up to 8,000 MT/s speed, 256GB/s bandwidth Storage - 2TB M.2 SSD Video Output HDMI 2.1b DisplayPort via USB-C port Networking 10GbE RJ45 port
OpenCV 5 release – New DNN engine with enhanced ONNX and LLM/VLM support, Intel, Arm, and RISC-V hardware optimizations
OpenCV 5 open-source computer vision library has recently been released with a brand-new DNN (Deep Neural Network) engine that provides better ONNX coverage and enables LLM/VLM support. The fifth version of the popular CV library also adds support for Intel, Arm, Qualcomm, and RISC-V hardware acceleration, improved 3D vision, and various new core features such as new data types, real N-dimensional and scalar support, and performance improvements. OpenCV 5's DNN Engine OpenCV 4.x supports about 22% of ONNX operators, and the new DNN engine in OpenCV 5 brings coverage to over 80%. That means models with dynamic shapes that used to fail on OpenCV 4.x, should now work, as the 5.x engine was rebuilt around a typed operation graph with proper shape inference, constant folding, and operator fusion. The table below shows the main difference between OpenCV 4.x and OpenCV 5 Since it's quite a big change, to make sure
Synaptics Astra SRW1500 Cortex-M52 Edge AI MCU features Ethos-U55 NPU, Wi-Fi 6/7, Bluetooth 6.0, 802.15.4 connectivity
Synaptics has introduced the Astra SRW1500 Series of AI-native Edge AI MCUs designed for smart IoT edge devices. The family features an Arm Cortex-M52 core, an optional Arm Ethos-U55 Neural Processing Unit (NPU), and tri-band Wi-Fi 7 or Wi-Fi 6, Bluetooth 6.0 LE, and 802.15.4 connectivity, in a single package. It targets low-latency on-device AI workloads such as voice-trigger detection, sound-event classification, and AI-enhanced Wi-Fi sensing. The series currently comes in four variants offering different wireless and USB options. Since sensing, AI processing, and wireless connectivity are integrated into a single chip, this approach helps reduce system costs, simplify PCB design, and lower power consumption in devices such as smart home hubs, building automation systems, and home appliances. Synaptics SRW1500 Series specifications: CPU Arm Cortex-M52 application core up to 200 MHz with Arm Helium (MVE) vector extensions for DSP and ML Arm Cortex-R4 WiFi core with 960 KB SRAM and 1408
WWDC26 starts today! I bet AI will be said at least 25 times during the Keynote.
What are your predictions?
#wwdc #wwdc26 #wwdc2026 #apple #ai #artificialintelligence #ArtificialintelligenceAI
cnx-software.com [Unofficial]
Homepage [Unofficial]
The New Yorker [Unofficial]
Mattiz6276 