NUMA emulation patch boosts Geekbench 6 benchmark results by up to 18% on Raspberry Pi 5

Raspberry Pi 5 NUMA emulation

Igalia Engineer, Tvrtko Ursulin has recently submitted a patch to the Linux kernel adding a NUMA (Non-Uniform Memory Access) emulation implementation for arm64 platforms that boosts the performance of 64-bit Arm targets by “splitting the physical RAM into chunks and utilizing an allocation policy to better utilize parallelism in physical memory chip organization”.

The NUMA emulation implementation was tested on a Raspberry Pi 5 SBC and the Geekbench 6 single-core score improved by 6%, while the multi-core score boosted by 18% after splitting into four emulated NUMA nodes. In other words, that’s like having a Broadcom BCM2712 CPU overclocked from 2.4 GHz up to 2.83 GHz.

Quectel FLM263D 2.4 GHz WiFi 6 and BLE 5.2 RISC-V module supports Amazon Alexa Connect Kit (ACK) SDK for Matter

Quectel FLM263D WiFi 6 Smart Home module

Quectel Wireless FLM263D is a standalone WiFi 6 and Bluetooth 5.2 LE module with a 320 MHz RISC-V microcontroller that supports Alexa Connect Kit (ACK) SDK for Matter for connectivity with Amazon Alexa, Google Home, Samsung SmartThings, Apple HomeKit, and other Matter-compliant Smart Home devices. The module also implements security standards such as secure boot and Mbed TLS encryption. The FLM263D offers certifications such as Works with Alexa (WWA) and supports features such as Matter Simple Setup (MSS) that allows zero-touch setup of Matter-compliant smart home devices through Alexa. Quectel FLM263D specifications: MCU – RISC-V wireless microcontroller up to 320 MHz, with 512KB SRAM and 4MB flash (likely Beken BK7235) Wi-Fi 6 & BLE 5.2 WLAN Protocol – IEEE 802.11b/g/n/ax Wi-Fi Frequency Band – 2.4 GHz Wi-Fi Modulation Mode – CCK, BPSK, QPSK, 16QAM, 64QAM Wi-Fi Operating Mode – AP, STA, STA + AP WiFi Rx sensitivity – Up to […]

Pine64 Oz64 RISC-V+Arm SBC to support NuttX RTOS and Debian Linux

Pine64 Oz64

Pine64 Oz64 is an upcoming credit card-sized SBC based on the SOPHGO SG2000 RISC-V+Arm(+8051) processor that currently runs NuTTX RTOS, and a Debian Linux image is also in the works. With a name likely inspired by the earlier Pine64 Ox RISC-V SBC, the Oz64 is a more powerful embedded board with 512 MB of DRAM integrated into the SG2000, a microSD card, an eMMC flash module connector, Ethernet port, WiFi 6 and Bluetooth 5.2, a USB 2.0 Type-A host port, and a 26-pin GPIO header. Pine64 Oz64 specifications: SoC – SOPHGO SG2000 Main core – 1 GHz 64-bit RISC-V C906 or Arm Cortex-A53 core (selectable) Minor core – 700 MHz 64-bit RISC-V C906 core Low-power core – 25 to 300 MHz 8051 MCU core with 8KB SRAM NPU – 0.5 TOPS INT8, supports BF16 Integrated 512MB DDR3 (SiP) Storage MicroSD card slot eMMC flash module connector Display – Optional 2-lane […]

GOWIN GW5AT-15 low-cost FPGA provides high-speed video interfaces up to 4K at 120 FPS

GOWIN GW5AT-15 FPGA

GOWIN Semiconductor has just announced the “Arora V” GW5AT-15 FPGA designed for high-speed video interfaces up to 4K at 120 FPS in space-constrained consumer and automotive applications. Manufactured with TSMC’s 22nm LP process, the new FPGA offers high-performance 12.5 Gbps SerDes, PCIe 3.0, MIPI CPHY, and DPHY interfaces. The GW5AT-15 FPGA also provides 15K LUTs of programmable logic, 118kbit of shadow SRAM, 630kbit of block SRAM (BSRAM), optional 64Mbit or 128Mbit of pseudo-SRAM (PSRAM), and optional 8Mbit of NOR Flash in a tiny 4.9mm x 5.3mm WLCSP package. Target applications include AR/VR headsets, 4K gaming tablets, and car infotainment systems. GOWIN GW5AT-15 FPGA specifications: FPGA Fabric 15120 logic elements (LUT4) 118kbit of shadow SRAM 630kbit of block SRAM (BSRAM) arranged as 35 x 18kbit 28x DSP slices, 12x DSP Lite slices  PSRAM GW5ART-15: 64Mbit (in MG132P package) or 128Mbit (in CM90P package) of pseudo SRAM (PSRAM) GW5ANRT-15: 64Mbit PSRAM DDR […]

SiFive announces Essential Gen4 RISC-V CPUs for embedded applications

SiFive Essential Gen4 RISC-V CPU IP

As the RISC-V Summit Europe 2024 is underway, SiFive has announced the fourth generation of its “Essential” RISC-V CPUs (Essential Gen4) with improved power efficiency and more flexible interfaces for SoCs used in embedded devices. The update covers 32-bit and 64-bit RISC-V cores including the U6 and U7-series 64-bit application processors, the S2, S6, and S7 64-bit real-time embedded processors, and the E2, E6, and E7 32-bit real-time embedded processors. Essential Gen4 IP keys features: Up to 40% runtime power reduction 8x different baseline embedded 32-bit and 64-bit cores From 2-stage single-issue to 8-stage superscalar Improved L2 cache and enhanced L1 memory Configuration and integration options CPU type, profile, and options On-chip memories selection System, peripheral, and front ports Advanced power management and security Debug and trace Software support includes embedded Linux and FreeRTOS operating systems and Eclipse-based IDE for C/C++ development. That’s another low-quality, light-on technical details announcement from […]

LILYGO T-Energy-S3 ESP32-S3 board offers 18650 battery holder, two 20-pin GPIO headers

ESP32-S3 board with 18650 battery holder

LILYGO T-Energy-S3 is an ESP32-S3 WiFi and Bluetooth IoT board with an 18650 battery holder, a USB-C port for charging and programming, and plenty of I/Os thanks to two 20-pin GPIO headers and a Qwiic connector. We’ve already covered a large number of ESP32-S3 boards and products, including some designs for battery operation such as the Tokay Lite AI camera board, LILYGO’s T7-S3 board, and Smart Bee Designs’ Bee S3 board among others. But that usually means providing a 2-pin connector and charging circuitry to connect a LiPo battery, but the T-Energy-S3 ESP32-S3 board works directly with an off-the-shelf 18650 rechargeable Lithium battery. T-Energy-S3 specifications: Wireless Module – ESP32-S3-WROOM-1-N16R8 Espressif Systems ESP32-S3R8 – Dual-core Tensilica LX7 @ up to 240 MHz with vector instructions for AI acceleration, 512KB RAM, 8MB PSRAM Storage – 16 MB flash Wireless – 2.4 GHz WiFi 4 and Bluetooth PCB antenna USB – 1x USB […]

Olimex NEO6502 is a W65C02 and Raspberry Pi RP2040-powered retro computer with HDMI, USB, and more

Olimex Neo6502 Dev Board

Olimex has recently launched NEO6502 an open-source hardware retro computer designed for educational and entertainment purposes, but that doesn’t mean it can’t be used for retro gaming. The board is very unique as it’s built around a 65C02 MPU and Raspberry Pi Pico RP2040 MCU in a dual-processor configuration. The MOS6502 handles the Apple II, Oric, and Commodore 64 emulators whereas the RP2040 handles everything else including HDMI (DVI) video output using the PicoDVI project. Previously we have written about similar retro computer kits such as the CERBERUS 2100, TinyLlama x86, DevTerm with ClockworkPi v3.14, Olimex AgonLight2, and many others feel free to check those out if you are interested in retro tech with a modern twist. Olimex NEO6502 board specifications Processors Western Design Center W65C02S 8-bit microprocessor @ 6.25 MHz with 64kB RAM, 32kB Graphics RAM Raspberry Pi RP2040 dual-core Cortex-M0+ MCU @ 133 MHz with 264 KB SRAM Storage […]

NDAA-compliant NVIDIA Jetson Orin NX and Nano bundles target drones

ARK Jetson Orin Nano NDAA Bundle

ARK Electronics has recently introduced NDAA-complaint NVIDIA Jetson Orin NX and Nano bundles designed for drones with the company’s Jetson PAB (Pixhawk Autopilot Bus) carrier board, an ARKV6X flight controller, and an SSD preloaded with NVIDIA Jetpack and the ARK UI. I’ve come across the board following a post on X pointing to an article on Tom’s Hardware about the US passing a bill called “Countering CCP Drones Act” that would effectively ban DJI drones in the US.  DJI is supposed to have a 90% market share of the U.S. hobby market, a 70% market share of the industrial market, and an over 80% market share of the first responder market, so not everybody will be happy if the bill passes in the senate and becomes enacted by the US president. But US companies making NDAA-compliant hardware and drones may bit mind, and that’s how I discovered ARK Electronics NVIDIA […]

EmbeddedTS embedded systems design