Just in time for Christmas, we're delighted to announce the release of two hotly anticipated products that we think will look great under the tree. One of them might even fit in a stocking if you push hard enough. Introducing Raspberry Pi 500, available now at $90, and the Raspberry Pi Monitor, on sale at $100: together, they're your complete Raspberry Pi desktop setup.

Integral calculus
Our original mission at Raspberry Pi was to put affordable, programmable personal computers in the hands of young people all over the world. And while we've taken some detours along the way - becoming one of the world's largest manufacturers of industrial and embedded computers - this mission remains at the heart of almost everything we do. It drives us to make lower-cost products like the $15 Raspberry Pi Zero 2 W, and more powerful products, like our flagship Raspberry Pi 5 SBC. These products provide just the essential processing element of a computer, which can be combined with the family television, and second-hand peripherals, to build a complete and cost-effective system.

Today we're happy to announce the much-anticipated launch of Raspberry Pi Compute Module 5, the modular version of our flagship Raspberry Pi 5 single-board computer, priced from just $45.

An unexpected journey
We founded the Raspberry Pi Foundation back in 2008 with a mission to give today's young people access to the sort of approachable, programmable, affordable computing experience that I benefitted from back in the 1980s. The Raspberry Pi computer was, in our minds, a spiritual successor to the BBC Micro, itself the product of the BBC's Computer Literacy Project. But just as the initially education-focused BBC Micro quickly found a place in the wider commercial computing marketplace, so Raspberry Pi became a platform around which countless companies, from startups to multi-billion-dollar corporations, chose to innovate. Today, between seventy and eighty percent of Raspberry Pi units go into industrial and embedded applications.

Details of what is claimed to be Google's upcoming Tensor G5 and G6 SoCs have popped up over on Notebookcheck.net and the site claims to have found the specs on a public platform, without going into any further details. Those that were betting on the Tensor G5—codenamed Laguna—delivering vastly improved performance over the Tensor G4, are likely to be disappointed, at least on the CPU side of things. As previous rumours have suggested, the chip is expected to be manufactured by TSMC, using its N3E process node, but the Tensor G5 will retain the single Arm Cortex-X4 core, although it will see a slight upgrade to five Cortex-A725 cores vs. the three Cortex-A720 cores of the Tensor G4. The G5 loses two Cortex-A520 cores in favour of the extra Cortex-A725 cores. The Cortex-X4 will also remain clocked at the same peak 3.1 GHz as that of the Tensor G4.

Interestingly it looks like Google will drop the Arm Mali GPU in favour of an Imagination Technologies DXT GPU, although the specs listed by Notebookcheck doesn't add up with any of the specs listed by Imagination Technologies. The G5 will continue to support 4x 16-bit LPDDR5 or LPDDR5X memory chips, but Google has added support for UFS 4.0 memory, something that's been a point of complaint for the Tensor G4. Other new additions is support for 10 Gbps USB 3.2 Gen 2 and PCI Express 4.0. Some improvements to the camera logic has also been made, with support for up to 200 Megapixel sensors or 108 Megapixels with zero shutter lag, but if Google will use such a camera or not is anyone's guess at this point in time.

Award-winning embedded solutions provider AAEON (stock code: 6579) has announced the launch of the RICO-MX8P, an NXP i.MX 8M Plus-powered fanless single-board built on the Pico-ITX Plus form factor. Utilizing the i.MX 8M Plus, the RICO-MX8P leverages a platform comprised of a quad-core Arm Cortex -A53 processor, a secondary core Arm Cortex M7, and an optional Neural Processing Unit (NPU) offering up to 2.3 TOPS of inference performance.

Equipped with an integrated Vivante GC7000 UltraLite 3D GPU, a dedicated VPU, MIPI DSI interface, and an HDMI 2.0 port, it is clear AAEON is positioning the RICO-MX8P as a candidate for multimedia applications, with digital signage, smart kiosk, and interactive Digital Out-of-Home (DOOH) advertising earmarked as potential uses. Further benefits to such use are evident from the board's support for APIs like OpenGL ES 3.1, Vulkan, and OpenCL 1.2, alongside its multiformat encoding and decoding capabilities.

Industry leader AAEON has expanded its RISC computing product portfolio with the release of the SRG-IMX8PL and PICO-IMX8PL, a Mini PC and 2.5" PICO-ITX board, respectively. Both products are powered by the NXP i.MX 8M Plus platform, featuring a quad-core Arm Cortex -A53 processor with a Neural Processing Unit (NPU) operating at up to 2.3 TOPS.

Built to provide cost-efficient IoT Gateway solutions in rugged environments, the SRG-IMX8PL and PICO-IMX8PL both offer wide temperature ranges of -40°C to 80°C with the use of a fanless heatsink, a 9 V to 36 V power input range. The SRG-IMX8PL Mini PC also features enhanced shock, drop, and vibration resistance. Dual LAN ports with IEEE 1588 and TSN capabilities, alongside Wi-Fi and 4G module support via M.2 2230 E-Key and full-size mini card, provide each device with broad connectivity options for industrial IoT use. Additionally, both the PICO-IMX8PL and SRG-IMX8PL support a wide range of operating systems, including Debian 11, Android 13, Windows 10 IoT, and Yocto, as well as data communication protocols such as Modbus, MQTT, and OPC Unified Architecture (OPC UA).

Arm has announced a pair of new Armv9 CPU cores today, alongside a refresh of a third. The new additions are the Cortex-X925—which is a huge model number jump from the previous Cortex-X4—and the Cortex-A725 which should be an upgraded Cortex-A720. Finally the Cortex-A520 has been refreshed to bring a 15 percent power efficiency improvement as well as support for 3 nm production nodes. Arm claims that the Cortex-A925 delivers its highest performance improvement ever over a previous generation with a single core uplift of up to 36 percent and an AI performance improvement of up to 46 percent compared to the Cortex-X4. The Cortex-X925 will support up to 3 MB private L2 cache and is tape-out ready for 3 nm production nodes.

The Cortex-A725 is said to offer a 35 percent performance efficiency improvement over the Cortex-A720 and it's been given performance boosts both when it comes to AI and gaming workloads. It's said to be up to 25 percent more power efficient than the Cortex-A720 and L3 cache traffic has been improved by up to 20 percent. Again, the Cortex-A720 is ready for production on a 3 nm node. Finally, Arm has also updated its DynamIQ Shared Unit to the DS-120 and here Arm has managed to reduce the typical workload power by up to 50 percent and the cache miss power by up to 60 percent. The DSU-120 scales to up to 14 Arm cores, suggesting that we might get to see some interesting new SoC implementations in the coming years from Arm's partners, although Arm's reference platform is a 2-4-2 configuration of the new cores.

Today we are thrilled to announce the latest innovation that promises to transform your gaming experience: Razer Cortex: Add-Ons. Our commitment at Razer has always been to push the boundaries of gaming technology and provide our users with the tools they need to succeed, whether they're casual gamers or competitive athletes. Today, we're taking another giant leap forward in fulfilling that promise, with the introduction of Cortex: Add-Ons - a new feature, crafted with the aim of maximizing your gaming experience.

What is Razer Cortex: Add-Ons?
Cortex: Add-Ons is your one-stop destination for all your gaming plug-in needs. We understand that gaming is not one-size-fits-all, which is why we're bringing you a platform that caters to your individual needs. Whether it's optimizing your performance, sharing your gaming highlights, or enriching your gaming sessions with unique tools, we've got you covered.

Radxa is a Chinese manufacturer of various Arm based devices and something of a minor competitor to the Raspberry Pi Foundation. The company has just launched its latest product which is called the ROCK 5 ITX. As the name implies, it's a Mini-ITX form factor motherboard, which in itself is rather unusual for Arm based hardware to start with. However, Radxa has designed the ROCK 5 ITX to be a NAS motherboard and this is the first time we've come across such a product, as most Arm based boards are either intended for hobby projects, software development or routers. This makes the ROCK 5 ITX quite unique, at least based on its form factor, as it'll be compatible with standard Mini-ITX chassis.

The SoC on the board is a Rockchip RK3588 which sports four Cortex-A76 cores at up to 2.4 GHz and four Cortex-A55 cores at 1.8 GHz. This is not exactly cutting edge, but should be plenty fast enough for a SATA drive based NAS. The board offers four SATA 6 Gbps connectors via an ASMedia ASM1164 controller, each with an individual power connector next to it. However, Radxa seems to have chosen to use fan-header type power connectors, which means it'll be hard to get replacement power cables. The board also has a PCIe 3.0 x2 M.2 slot for an NVMe drive. The OS boots from eMMC and Radxa supports its own Roobi OS which is Debian Linux based.

ScaleFlux, a leader in deploying computational storage at scale, today announced its commitment to integrating the Arm Cortex -R82 processor in its forthcoming line of enterprise Solid State Drive (SSD) controllers. The Cortex-R82, is the highest performance real-time processor from Arm and the first to implement the 64-bit Armv8-R AArch64 architecture, representing a significant advancement in processing power and efficiency for enterprise storage solutions.

ScaleFlux's adoption of the Cortex-R82 is a strategic move to leverage the processor's high performance and energy efficiency. This collaboration underscores ScaleFlux's dedication to delivering cutting-edge technology in its SSD controllers, enhancing data processing capabilities and efficiency for data center and AI infrastructure worldwide.

Samsung and ARM announced a collaborative project last week—the partners are aiming to deliver an "optimized next generation Arm Cortex -X CPU" developed on the latest Gate-All-Around (GAA) process technology. Semiconductor industry watchdogs believe that Samsung Foundry's 3 nm GAA process did not meet sales expectations—reports suggest that many clients decided to pursue advanced three nanometer service options chez TSMC. The South Korean multinational manufacturing conglomerate is setting its sights forward—with an in-progress SF2 GAAFET process in the pipeline—industry insiders reckon that Samsung leadership is hoping to score a major victory within this next-gen market segment.

Lately, important industry figures have been hyping up Backside Power Supply Delivery Network (BSPDN) technology—recent Intel Foundry Services (IFS) press material lays claim to several technological innovations. A prime example being an ambitious five-nodes-in-four-years (5N4Y) process roadmap that: "remains on track and will deliver the industry's first backside power solution." A Chosun Business report proposes that Samsung is working on Backside Power Supply designs—a possible "game changer" when combined with in-house 2 nm SF2 GAAFET. Early experiments, allegedly, involving two unidentified ARM cores have exceeded expectations—according to Chosun's sources, engineers were able to: "reduce the chip area by 10% and 19%, respectively, and succeeded in improving chip performance and frequency efficiency to a single-digit level." Samsung Foundry could be adjusting its mass production timetables, based on freshly reported technological breakthroughs—SF2 GAAFET + BSPDN designs could arrive before the original targeted year of 2027. Prior to the latest developments, Samsung's BSPDN tech was linked to a futuristic 1.7 nm line.

Fibocom, a global leading provider of IoT (Internet of Things) wireless solutions and wireless communication modules, launches a new series of 5G RedCap module integrated with MediaTek's T300 5G modem, which is the world's first 6 nm radio frequency system-on-chip (RFSOC) single die solution for 5G RedCap. By integrating a single-core Arm Cortex-A3 processor in a significantly compact PCB area, the FM330 series are optimal solutions that offer extended coverage, increased network efficiency and device battery life for industry customers.

Compliant with 3GPP R17 standards, the FM330 series supports mainstream 5G frequency bands worldwide and is capable of reaching a maximum bandwidth of 20 MHz, thus ensuring the peak data rate of up to 227 Mbps downlink and 122 Mbps uplink, sufficient to meet the demand for 5G applications with less data throughput while balancing the power efficiency. In hardware design, it adopts the M.2form factor measured at 30x42mm benefiting from the unique RFSOC solution integrated with T300, in addition to the optimized antenna design in 1T2R, which significantly saves the PCB area. Moreover, FM330 series is pin-compatible with Fibocom LTE Cat 6 module FM101, easing the concerns for customers' migration from 4G to 5G. Furthermore, the module provides 64QAM/256QAM (optional) modulation scheme to greatly optimize the cost and size.

Samsung Electronics Co., Ltd., a world leader in advanced semiconductor technology, today announced a collaboration to deliver optimized next generation Arm Cortex -X CPU developed on Samsung Foundry's latest Gate-All-Around (GAA) process technology. This initiative is built on years of partnership with millions of devices shipped with Arm CPU intellectual property (IP) on various process nodes offered by Samsung Foundry.

This collaboration sets the stage for a series of announcements and planned innovation between Samsung and Arm. The companies have bold plans to reinvent 2-nanometer (nm) GAA for next-generation data center and infrastructure custom silicon, and a groundbreaking AI chiplet solution that will revolutionize the future generative artificial intelligence (AI) mobile computing market.

MediaTek, one of the first adopters of Wi-Fi 7 technology, now has the industry's most comprehensive Wi-Fi 7 portfolio with today's introduction of the company's new Filogic 860 and Filogic 360 solutions. Together, these second-generation additions aim to further expand MediaTek's platform of cutting-edge products that utilize the latest technology advancements in connectivity while achieving peak performance and always-on reliability.

Filogic 860 combines a Wi-Fi 7 dual-band access point with a new advanced network processor solution and is ideal for enterprise access points, service provider Ethernet gateways and mesh nodes, as well as retail and IoT router applications. Filogic 360 is a stand-alone client solution that integrates Wi-Fi 7 2x2 and dual Bluetooth 5.4 radios in a single chip, and is designed to deliver next-generation Wi-Fi 7 connectivity to edge devices, streaming devices and a vast array of other consumer electronics.

MediaTek today announced the Dimensity 9300, its newest flagship mobile chip with a one-of-a-kind All Big Core design. The unique configuration combines extreme performance with MediaTek's industry-leading power efficiency to deliver unmatched user experiences in gaming, video capture and on-device generative AI processing.

"The Dimensity 9300 is MediaTek's most powerful flagship chip yet, bringing a huge boost in raw computing power to flagship smartphones with our groundbreaking All Big Core design," said Joe Chen, President at MediaTek. "This unique architecture, combined with our upgraded on-chip AI Processing Unit, will usher in a new era of generative AI applications as developers push the limits with edge AI and hybrid AI computing capabilities."

Synopsys today announced it has expanded its collaboration with Arm to provide optimized IP and EDA solutions for the newest Arm technology, including the Arm Neoverse V2 platform and Arm Neoverse Compute Subsystem (CSS). Synopsys has joined Arm Total Design where Synopsys will leverage their deep design expertise, the Synopsys.ai full-stack AI-driven EDA suite, and Synopsys Interface, Security, and Silicon Lifecycle Management IP to help mutual customers speed development of their Arm-based CSS solutions. The expanded partnership builds on three decades of collaboration to enable mutual customers to quickly develop specialized silicon at lower cost, with less risk and faster time to market.

"With Arm Total Design, our aim is to enable rapid innovation on Arm Neoverse CSS and engage critical ecosystem expertise at every stage of SoC development," said Mohamed Awad, senior vice president and general manager, Infrastructure Line of Business at Arm. "Our deep technical collaboration with Synopsys to deliver pre-integrated and validated IP and EDA tools will help our mutual customers address the industry's most complex computing challenges with specialized compute."

It has been over four years since the release of the Raspberry Pi 4, and in that time a lot has changed in the maker board and single-board computer landscape. For the Raspberry Pi Foundation there were struggles with worldwide demand and production capacity brought on by the global pandemic starting in 2020, and plenty of new competitors came to the scene to offer ready to order alternatives to the venerable RPi 4. Today however the production woes have been assuaged and a new generation of Raspberry Pi is here; the Raspberry Pi 5.

Raspberry Pi 5 is being announced in advance of availability unlike every prior RPi device launch. Pre-orders are open with many of the listed Approved Resellers on RPi's website starting today but unit shipments aren't expected until near the end of October 2023. As part of this pre-order scheme, RPi Foundation is withholding pre-orders from bulk customers and will be dealing in single-unit sales for individuals until at least the end of the year, as well as running some promotions with The MagPi and HackSpace magazines to give priority access to their subscribers. Genuinely nice to see, considering how hard it was to obtain a Pi 4 for the average Joe over the last couple years. The two announced prices for the RPi 5 are $60 USD for the 4 GB variant, and $80 USD for the 8 GB variant; or about $5 USD more than current reseller pricing on comparable configurations of the Raspberry Pi 4.

Useful Sensors, an AI-focused start-up, today launched the world's first low-cost, off-the-shelf AI module to enable intuitive, natural language interaction with electronic devices, locally and privately, with no need for an account or internet connection. The new AI-In-A-Box module can answer queries and solve problems in a way similar to well-known AI tools based on a large language model (LLM). But thanks to compression and acceleration technologies developed by Useful Sensors, the module hosts its LLM file locally, enabling its low-cost microprocessor to understand and respond instantly to spoken natural language queries or commands without reference to a data center.

Disconnected from the internet, the AI-In-A-Box module definitively eliminates user concerns about privacy, snooping, or dependence on third-party cloud services that are prevalent with conventional LLM-based AI products and services marketed by large technology companies. The AI-In-A-Box module is available to buy now at CrowdSupply, priced at $299.

The revolutionary UNO R4, announced on Arduino Day, is now available on the Arduino Store! The fourth version of the iconic, beloved UNO adds a whole new dimension to the world of DIY and making revolutionized by the simple 8-bit microcontroller over a decade ago. Take your maker potential to new heights: UNO R4 has a speedy 32-bit Arm Cortex-M4 and offers a 16-fold increase in memory, as well as more connectors and connectivity options than ever - in two variants: UNO R4 Minima and UNO R4 WiFi.

Both variants allow you to start making or easily upgrade UNO R3-based projects with more computational power, memory, and speed than previous versions - courtesy of the RA4M1 microcontroller by Renesas - while maintaining the same form factor and 5 V operating voltage.

Mobile devices touch every aspect of our digital lives. In the palm of your hand is the ability to both create and consume increasingly immersive, AI-accelerated experiences that continue to drive the need for more compute. Arm is at the heart of many of these, bringing unlimited delight, productivity and success to more people than ever. Every year we build foundational platforms designed to meet these increasing compute demands, with a relentless focus on high performance and efficiency. Working closely with our broader ecosystem, we're delivering the performance, efficiency and intelligence needed on every generation of consumer device to expand our digital lifestyles.

Today we are announcing Arm Total Compute Solutions 2023 (TCS23), which will be the platform for mobile computing, offering our best ever premium solution for smartphones. TCS23 delivers a complete package of the latest IP designed and optimized for specific workloads to work seamlessly together as a complete system. This includes a new world-class Arm Immortalis GPU based on our brand-new 5th Generation GPU architecture for ultimate visual experiences, a new cluster of Armv9 CPUs that continue our performance leadership for next-gen artificial intelligence (AI), and new enhancements to deliver more accessible software for the millions of Arm developers.

Early details of MediaTek's next generation mobile chipset have emerged this week, courtesy of renowned leaker Digital Chat Station via their blog on Weibo. The successor to MediaTek's current flagship Dimensity 9200 mobile chipset will likely be called "Dimensity 9300" - a very imaginative bump up in numbering - with smartphone brand Vivo involved as a collaborator. The tipster thinks that the fabless semiconductor company has contracted with TSMC for fabrication of the Dimensity 9300 chipset - and the foundry's N4P process has been selected by MediaTek, which could provide a bump in generational performance when compared to the older 4 nm and 5 nm standards used for past Dimensity SoC ranges. It should be noted that the current generation Dimensity 9200 chipset is presently manufactured via TSMC's N4P process.

MediaTek is seeking to turnaround its fortunes in the area of flagship mobile chipsets - industry watchdogs have cited a limited uptake of the Taiwanese company's Dimensity 9200 SoC as a motivating factor in the creation of a very powerful successor. Digital Chat Station suggests that the upcoming 9300 model will pack enough of a hardware punch to rival Qualcomm's forthcoming Snapdragon 8 Gen 3 SoC - both chipsets are touted to release within the same time period of late 2023. According to previous speculation, Qualcomm has also contracted with TSMC's factory to pump out the Snapdragon 8 Gen 3 via the N4P (4 nm) process.

Softbank, the owner of Arm Ltd., is preparing everything it can to ensure a successful initial public offering (IPO) of Arm. However, ahead of the IPO, we have more information about Arm's plans to change its licensing and pricing structures to collect more royalties and ensure higher cash flow for future investors. Currently, Arm licenses technology in the form of intellectual property (IP), usually in different flavors of Cortex-A CPU cores that go inside processors for phones and laptops. Chipmakers that use the IP have additional expenses such as Arm ISA license fee and per-chip royalty, which is based on the chip's average selling price.

However, according to Financial Times, we have a new pricing structure that changes how Arm bills its partners and customers. From now on, Arm will grant licenses to chipmakers and ask them to only ship to device makers with an agreement with Arm. Additionally, these device makers now pay per-device royalty based on the device's average selling price (ASP). This ensures that Arm's fee applies to the higher margin product, which means that ultimately Arm will collect more cash flow from its customers and partners. Currently, the old model charges around 1-2 percents per chip in each smartphone, considering the ASP of smartphone chips to be $40 for Qualcomm, $17 for MediaTek, and $6 for Unisoc. However, taking the ASP of a mobile phone at $335, as recorded in 2022, the fee would be much higher. People familiar with the matter noted that Arm will apply this pricing structure as early as 2024. Apple and Samsung are not impacted by this change, as both companies enjoy their own agreements with Arm.

Intel has quietly canceled its Thunder Bay hybrid system-on-chip (SoC) that combines standard general-purpose CPU Cores and Movidius Vision Processing Unit (VPU) cores. Such chips were aimed at commercial and Internet-of-Things (IoT) applications that relied on computer vision acceleration and edge-computing applications. According to the latest report, it appears that Thunder Bay is canceled as Intel has released a set of patches that removed the Thunder Bay code from the Linux kernel.

Intel kept Thunder bay details well hidden and while earlier rumors pointed to a combination of Intel Xeon x86 CPU cores and Movidius VPU cores, the only Thunder Bay support in Linux patches where showing a combination of ARM Cortex-A53 low-power cores with the Movidius VPU. Intel acquired Movidius back in 2016 and while it is not something that Intel talks about often, there are several products based on Movidius VPUs, including Neural Compute Stick, Intel drones, the Intel RealSense Tracking Camera, and most recently, the Intel Movidius 3700VC VPU, formerly Keem Bay. In any case, it appears that Intel is abandoning an idea of combining general purpose x86 CPU cores and Movidius VPU cores.

Kernel information for a chipset code-named 'Quadra' has been leaked by Home IT. It appears that this SoC is under development at Samsung Electronics, and could be featured in the company's next generation flagship smartphone range. The recently released Galaxy S23 smartphone series is powered by Qualcomm's cutting-edge Snapdragon 8 Gen 2 processor, which ended up being a fine choice for the end user. This was not an ideal partnership for the semiconductor giant, considering its constant push to promote internally developed hardware.

Industry experts have praised Samsung for not integrating an Exynos processor into this generation of devices, including regional model variations, but that has not stopped the company's persistent development cycle of proprietary mobile CPUs. Rumors point to a collaboration with Google, and the Exynos 2300 SoC looks to form the basis of the latter's next generation Tensor G3 chipset. The leak reveals that the standard 2300 has been designed as a 1+4+4 core configuration, comprised of four performance cores, four efficiency cores, and one super core.

MediaTek, a fabless semiconductor company powering two billion devices a year, and Kontron, a leading global provider of IoT and Embedded Computer Technology (ECT), today announced that Kontron is integrating MediaTek's Genio 1200 and Genio 700 chipsets into its new modules for IoT applications.

MediaTek Genio is a complete platform stack for IoT, offering brands powerful and efficient chipsets, open platform software development kits (SDK), and a developer portal with comprehensive resources and tools. Genio chipsets are ideal for a wide variety of commercial, industrial, and retail applications that require highly responsive processing, advanced multimedia support, and multi-tasking.

Renesas Electronics Corporation, a premier supplier of advanced semiconductor solutions, today announced that it will present the first live demonstrations of artificial intelligence (AI) and machine learning (ML) implementations on an MCU based on the Arm Cortex -M85 processor. The demos will show the performance uplift in AI/ML applications made possible by the new Cortex-M85 core and Arm's Helium technology. They will take place in the Renesas stand - Hall 1, Stand 234 (1-234) at the embedded world 2023 Exhibition and Conference in Nuremburg, Germany from March 14-16.

At embedded world in 2022, Renesas became the first company to demonstrate working silicon based on the Arm Cortex-M85 processor. This year, Renesas is extending its leadership by showcasing the features of the new processor in demanding AI use cases. The first demonstration showcases a people detection application developed in collaboration with Plumerai, a leader in Vision AI, that identifies and tracks persons in the camera frame in varying lighting and environmental conditions. The compact and efficient TinyML models used in this application lead to low-cost and lower power AI solutions for a wide range of IoT implementations. The second demo showcases a motor control predictive maintenance use case with an AI-based unbalanced load detection application using Tensorflow Lite for Microcontrollers with CMSIS-NN.