Tag: RISC-V

  • Prototype 2T1R Control Chip for Neuromorphic Computing Developed

    Prototype 2T1R Control Chip for Neuromorphic Computing Developed

    Key Takeaways

    1. In-memory computing is changing computer architecture by moving processing closer to memory for improved efficiency.
    2. A new 2T1R memristor design enhances energy efficiency for AI and edge devices by minimizing sneak path currents and leakage.
    3. The design supports analogue vector-matrix multiplication (VMM), crucial for machine learning applications.
    4. It effectively addresses virtual ground issues and wire resistance to boost performance and reduce power usage.
    5. This technology paves the way for faster, more integrated AI operations in memory, hinting at a future where hard drives could process information more intelligently.

    What if your hard drive did more than just store your files? Imagine it could think and react to information right where it’s located. This idea is part of in-memory computing, which is changing the way computers are built by moving processing closer to memory to improve efficiency.

    New Memristor Design

    Researchers from Forschungszentrum Jülich and the University of Duisburg-Essen have introduced a fresh design based on a 2T1R memristor. This innovation aims to facilitate a more energy-efficient approach for AI and edge devices.

    The details of this design, shared on arXiv, include the integration of two transistors and one memristor in each cell, which helps regulate current to minimize sneak path currents—an issue often faced in memristor arrays. Unike regular memory, this new design connects both memristor terminals to ground when not in active use. This method could enhance signal stability and cut down on leakage.

    Enhancing Machine Learning

    The architecture is meant to enable analogue vector-matrix multiplication (VMM), a key element in machine learning. It achieves this by managing memristor conductance with built-in DACs, PWM signals, and controlled current paths. A functional 2×2 test array was successfully created using standard 28 nm CMOS technology.

    By tackling virtual ground concerns and the impact of wire resistance, the design seeks to boost performance reliability and decrease power usage. Compatible with RISC-V control and digital connections, the 2T1R structure may set the stage for scalable neuromorphic chips, allowing for speedier and more compact AI acceleration directly in memory.

    Future of AI and Memory

    While your hard drive isn’t quite thinking yet, the technology that could make this a reality is already being developed in silicon. This progression hints at a future where AI operates faster and more seamlessly integrated with memory.

    Source:
    Link

    Tags: , ,
  • StarPro64: Powerful Raspberry Pi Alternative with NPU and PCIe

    StarPro64: Powerful Raspberry Pi Alternative with NPU and PCIe

    Key Takeaways

    1. The StarPro64 is a new developer board from Pine64, priced at $250 but currently out of stock.
    2. It features an ESWIN EIC7700X processor with four SiFive P550 cores based on RISC-V architecture, making it a competitor to the Raspberry Pi 5.
    3. The board includes a high-performance NPU capable of 19.95 TOPS for local AI applications, such as surveillance tasks.
    4. Key specifications include 32 GB of LPDDR5 RAM, 128Mb SPI boot flash, and options for eMMC storage and microSD booting.
    5. Connectivity features consist of PCIe 3, multiple USB ports, 4K HDMI output, dual Gigabit Ethernet ports, WiFi 6, and Bluetooth 5.3.

    The StarPro64 has been officially introduced – or at least it is now featured on Pine64’s online shop. The listed price is $250, but currently, the board seems to be “out of stock,” indicating that it’s sold out at this time.

    Developer Board Details

    The StarPro64 serves as a developer board and is basically a competitor to the Raspberry Pi 5. However, it’s equipped with a different processor architecture, specifically an ESWIN EIC7700X, which includes four SiFive P550 cores based on RISC-V architecture. One notable feature is its high-performance NPU, which reportedly delivers INT8 performance of 19.95 TOPS per second, making it quite powerful for running local AI applications. A realistic application could be in the field of surveillance cameras, particularly for face or license plate recognition.

    Specifications and Features

    Standard features include 32 GB of LPDDR5 RAM and 128Mb SPI boot flash. Users have the option to add an eMMC module (up to 256GB) and utilize a microSD slot for booting purposes. For connectivity, it offers a four-lane PCIe 3 connector, two USB 3.0 ports, and two USB 2.0 ports. Additionally, there are MIPI DSI and MIPI CSI interfaces for connecting displays and cameras. It also supports HDMI output at 4K resolution and 60 Hz. Furthermore, network capabilities include two Gigabit Ethernet ports alongside dual-band WiFi 6 and Bluetooth 5.3.

    Conclusion

    Pine64’s new offering, the StarPro64, appears to be a compelling choice for developers looking for advanced features and performance. However, potential buyers might need to wait until it’s back in stock to secure their own unit.

    Source:
    Link

    Tags: , ,
  • Alpha-One Mini PC with NPU and 20 TOPS Performance

    Alpha-One Mini PC with NPU and 20 TOPS Performance

    Key Takeaways

    1. Most AI models are cloud-based, which can create legal issues with sensitive data.
    2. AMD and Intel are integrating NPUs into their chipsets, while more devices emphasize local AI capabilities.
    3. The Pine64 Alpha-One mini PC features a robust NPU with 20 TOPS INT8 performance, based on RISC-V architecture.
    4. It builds on the StarPro64 SBC and includes 64GB of eMMC storage and a preloaded 7b Deepseek/Oen LLM.
    5. The device operates quietly with passive cooling, uses less than 30 watts, supports 4K monitors, and is priced at $330.

    The majority of AI models today operate in the cloud, relying on external computers. While this is easy to use, it can lead to (legal) complications when handling sensitive data for analysis. Even though regular users typically do not run AI models on their own devices, companies such as AMD and Intel are already incorporating NPUs into their existing chipsets. Additionally, some network storage devices and security cameras are beginning to emphasize local AI capabilities.

    New Mini PC with RISC-V Architecture

    Recently, a new mini PC utilizing the RISC-V architecture and boasting a robust NPU has been introduced. The Pine64 Alpha-One features an NPU with an INT8 performance reaching 20 TOPS. It’s important to mention that other contemporary APUs could be considerably more powerful in this area.

    Based on Existing Technology

    The Pine64 device is not completely novel, as it builds off the StarPro64 SBC. This board is equipped with an ESWIN EIC7700X SoC which has four SiFive 550 processing cores that can operate at speeds of up to 1.4GHz. In addition, the Alpha-One includes 64GB of eMMC storage, and comes preloaded with 7b Deepseek/Oen LLM in Docker containers. The manufacturer also asserts that it achieves a throughput of about 3.5 tokens per second (for a 7b LLM), which should suffice for many novices eager to run AI models locally.

    Power and Connectivity Features

    Power is delivered through a DC barrel plug at 12 volts, and USB Type-C PD is not available. This mini PC has a passive cooling system, ensuring that there’s no fan noise, as the SoC’s temperatures are regulated by a heat pipe. The device operates on less than 30 watts and provides various connectivity options like Ethernet, USB 3.0, and USB 2.0. Furthermore, the Pine64 Alpha-One is capable of supporting 4K monitors at resolutions up to 60Hz, and PCIe Gen 3 with four lanes can be utilized for specific expansion modules. The new mini PC is priced at $330, yet no specific release date has been announced yet.

    Source:
    Link

    Tags: , ,
  • Muse Pi Pro: RISC-V SoC, M.2 & PCIe Raspberry Pi Alternative

    Muse Pi Pro: RISC-V SoC, M.2 & PCIe Raspberry Pi Alternative

    Key Takeaways

    1. The Muse Pi Pro uses a RISC-V processor (SpacemiT M1) with 8 CPU cores and an NPU, unlike the Raspberry Pi which uses ARM architecture.
    2. It offers memory options of 8GB or 16GB LPDDR4X RAM and storage options of 32GB or 64GB eMMC, along with support for M.2 SSDs.
    3. Connectivity features include Gigabit Ethernet, WiFi 6, Bluetooth 5.2, and a mini PCIe slot for a cellular modem.
    4. The board has four USB 3.0 Type-A ports, one USB 2.0 Type-C port, and supports MIPI-CSI and MIPI-CSO for displays and cameras.
    5. The Muse Pi Pro is compatible with various Linux distributions, including Ubuntu, and was initially priced around $120.

    The Muse Pi Pro is a fresh single-board computer that resembles the popular Raspberry Pi and might be a good substitute for it. But, there is a key difference that sets them apart. Unlike the Raspberry Pi which uses ARM architecture, the Muse Pi Pro runs on a RISC-V processor. Specifically, this board is driven by the SpacemiT M1, featuring 8 CPU cores and a Neural Processing Unit (NPU) that delivers a performance of 2 TOPS, making it ideal for AI tasks.

    Memory and Storage Options

    Depending on which model you choose, the Muse Pi Pro comes with either 8GB or 16GB of LPDDR4X RAM. Storage options include eMMC 4.1 with either 32GB or 64GB capacities. This compact computer also supports an M.2 2230 SSD, and you can connect a cellular modem using the mini PCIe 2.1 slot, which has one lane. Additionally, the Muse Pi Pro is equipped with Gigabit Ethernet, WiFi 6, and Bluetooth 5.2 for connectivity. Dimensions for the board are around 3.3 x 2.2 inches, and it includes a 40-pin GPIO header for linking sensors and actuators.

    Ports and Compatibility

    There are four USB 3.0 Type-A ports on the Muse Pi Pro, along with one USB 2.0 Type-C port. You can connect displays and cameras via MIPI-CSI and MIPI-CSO interfaces. The single-board computer supports a range of Linux distributions, including Ubuntu, and it receives power through the USB Type-C connection. The Muse Pi Pro was briefly available on AliExpress with prices starting at about $120, but that listing has been removed since then.

    Source:
    Link

    Muse Pi Pro: RISC-V SoC, M.2 & PCIe Raspberry Pi Alternative

     

    Tags: , ,
  • Orange Pi RV2: New RISC-V SBC with PCIe 2.0 and Gigabit Ethernet

    Orange Pi RV2: New RISC-V SBC with PCIe 2.0 and Gigabit Ethernet

    Key Takeaways

    1. Orange Pi has released the new Orange Pi RV2, a successor to the previous RV model.
    2. The RV2 features a new octa-core Ky X1 RISC-V chipset with an AI accelerator, improving performance.
    3. The board includes a second Gigabit Ethernet port and M.2 connectors for PCIe 2.0 x2 support.
    4. Pricing for the RV2 starts at $29.50 for the 2 GB version, with options for 4 GB and 8 GB at higher prices.
    5. Key ports on the RV2 include HDMI, USB Type-A, MIPI-CSI connectors, and a 26-pin expansion header.

    Orange Pi has had a rather subdued beginning to 2024, rolling out several single-board computers (SBCs) this year. Some of the announcements from 2024, like the RISC-V based Orange Pi RV, have yet to move into production.

    New Arrival

    However, the company has now launched a sort of successor that is already available worldwide. At first glance, the Orange Pi RV2 resembles last year’s Orange Pi RV quite a bit. Yet, this new SBC has a second Gigabit Ethernet port added to its compact 89 x 56 mm size, along with a couple of M.2 connectors that can support up to PCIe 2.0 x2.

    Key Features

    One of the most significant upgrades is the introduction of a new RISC-V chipset. To give some context, the original RV was announced with a quad-core StarFive JH7110. In comparison, the RV2 comes with an octa-core Ky X1 chipset, which boasts an AI accelerator capable of delivering up to 2 TOPS of theoretical performance.

    In addition, the RV2 is equipped with HDMI and USB Type-A ports, as well as MIPI-CSI connectors. Orange Pi has also added a 26-pin expansion header and a MicroSD card reader. Currently, you can order the Orange Pi RV2 for around $29.50 with 2 GB of LPDDR4X RAM on AliExpress; curiously, prices on Amazon are higher. There are also 4 GB and 8 GB versions available for $39.50 and $49.50, respectively. For more information, visit Orange Pi’s official site.

    Source:
    Link

    Orange Pi RV2: New RISC-V SBC with PCIe 2.0 and Gigabit EthernetOrange Pi RV2: New RISC-V SBC with PCIe 2.0 and Gigabit Ethernet
    Tags: , ,
  • Encouraging Local Use of Open-Source RISC-V Chips in China

    Encouraging Local Use of Open-Source RISC-V Chips in China

    Key Takeaways

    1. The XiangShan project, initiated by China’s Academy of Sciences, aims to launch an open-source chip by the end of 2023.
    2. Chinese officials are developing policy guidance to promote the use of RISC-V chips, with documentation being worked on by eight government agencies.
    3. Major Chinese companies are increasingly adopting RISC-V architecture due to its cost-effectiveness and geopolitical neutrality.
    4. The chip market is dominated by American companies like Intel and AMD, which use proprietary technologies.
    5. The largest providers of RISC-V intellectual property in China include Alibaba-owned XuanTie and Nuclei System Technology, with growing demand expected for these chips.

    Back in 2019, the XiangShan project was started by China’s Academy of Sciences, aiming to introduce an open-source chip of the same name by the end of this year. A recent update has shown consistent progress, and insiders who know about local laws have reported that Chinese officials are developing policy guidance to promote the usage of RISC-V chips. Although many specifics are still vague, it appears that the draft could be wrapped up this month.

    Policy Development Details

    As reported by Reuters, two unnamed sources discussed the topic, revealing that eight government agencies are currently working on the documentation. These agencies include the Cyberspace Administration, the Ministry of Industry and Information Technology, the Ministry of Science and Technology, along with the National Intellectual Property Administration. Unfortunately, these four departments have not provided any comments about the ongoing policy discussions, but we will continue monitoring this situation closely.

    Industry Context

    The chip market is mainly controlled by American firms like Intel and AMD, both using proprietary technologies. In contrast, the open-source RISC-V chips have a geopolitically neutral stance. Furthermore, Chinese businesses have been increasingly focusing on this architecture due to its cost-effectiveness. The largest for-profit providers of RISC-V intellectual property in China are Alibaba-owned XuanTie and a startup named Nuclei System Technology. Because RISC-V chips generally require less computing power compared to alternative solutions, applications based on DeepSeek could drive higher demand for these chips both in China and internationally.

    Source:
    Link

    Tags: , ,
  • China’s RISC-V Processor Project Aims for 2025 Launch with High Goals

    China’s RISC-V Processor Project Aims for 2025 Launch with High Goals

    The Chinese Academy of Sciences (CAS) is set to introduce its XiangShan open-source processor in 2025. This development represents a significant step in China’s efforts to create more domestic chips. Bao Yungang, the deputy director at the Institute of Computing Technology, shared this timeline in a recent Weibo post, emphasizing their steady progress in high-performance computing.

    Progress on XiangShan Initiative

    The XiangShan project was initiated by CAS in 2019, and it has achieved notable advancements with its third-generation architecture, named Kunminghu. Leaders of the project claim that this new architecture can offer performance within approximately 8 percent of Arm’s Neoverse N2 CPU core. This puts it close to being suitable for applications in cloud computing, high-performance computing (HPC) systems, and machine learning tasks.

    Focus on Optimization

    In 2024, the development team concentrated on enhancing Kunminghu’s efficiency regarding area and power usage. The processor features several parallel processing units that handle integer, floating-point, and vector operations, along with out-of-order execution capabilities. Simulations indicate that it can operate at speeds reaching up to 3GHz.

    Comparison with Previous Chips

    The earlier version, known as the Nanhu chip, was manufactured using a 14nm process node and could reach clock speeds of up to 2GHz. Launched in late 2023, Nanhu represented the project’s second generation of silicon technology.

    RISC-V and Licensing

    XiangShan utilizes the RISC-V instruction set licensed under the Mulan PSL-2.0, which grants users royalty-free rights to modify and use the design. This approach contrasts with the traditional licensing models provided by major companies like Arm and could potentially reshape the future of processor design, much like how Linux transformed the industry.

    Increasing Scrutiny

    Despite the advancements, the XiangShan project is facing increased scrutiny from the United States as China continues to embrace RISC-V technology. Recently, the U.S. Commerce Department has been evaluating possible security issues. Google has even taken steps to remove RISC-V support from the Android kernel, which could pose challenges for any upcoming RISC-V-based smartphones.

    Source:
    Link

    Tags: , ,
  • Samsung Brings Tizen OS to Open-Source RISC-V Architecture

    Samsung Brings Tizen OS to Open-Source RISC-V Architecture

    Samsung has adapted the Tizen OS for the open-source RISC-V CPU architecture. This move allows developers to get closer to launching Tizen smart devices that utilize royalty-free CPUs.

    Understanding CPU Architectures

    Typical CPU architectures found in computers and smart devices from major companies like Apple, AMD, ARM, and Intel are burdened with numerous patents that restrict royalty-free, open-source usage. Companies that produce compatible CPUs are required to pay licensing fees, which raise the overall costs of smart devices. This system also obstructs the ability for others to create their own versions freely. In contrast, the RISC-V royalty-free open-source architecture removes these licensing fees and development barriers.

    For readers interested in exploring RISC-V software development without spending much, the Raspberry Pi Pico 2 is available for purchase on Amazon.

    Samsung’s Engagement with RISC-V

    In 2023, Samsung became part of the RISE (RISC-V Software Ecosystem) project to help enhance the RISC-V software landscape. The announcement regarding Tizen OS for RISC-V was made on October 4 during Samsung’s annual Developer Conference (SDC). Additionally, the company revealed that it has also ported Flutter, an open-source UI framework created by Google, to Tizen OS on RISC-V. A developer SDK for Tizen OS is anticipated to be released in 2026.

    Highlights from the Samsung Developer Conference

    On October 4, 2024, Samsung Electronics held its yearly Developer Conference (SDC) at the San Jose McEnery Convention Center. This event marked ten years of teamwork and advancement with developers, during which the company outlined its goal of delivering a more tailored and secure experience through the latest multi-device AI technologies under the theme “AI for All.”

    With approximately 3,000 developers, partners, and media present, Jong-Hee (JH) Han, Vice Chairman, CEO and Head of Device eXperience (DX) Division at Samsung Electronics, stated, “We have continued our journey of innovation with our many developers and partners for the last ten years,” emphasizing the need for open collaboration to enhance competitiveness and foster mutual growth in the current AI-driven landscape.

    Future Vision and Commitment

    JH provided more insights into the company’s dedication to utilizing AI technology for improved security, enhanced personalization, better family care experiences, and new opportunities in the B2B sector. “Samsung Electronics will keep pushing AI technology to a point where it automatically recognizes each user—based on their voice and location—creating a more personalized interaction,” he clarified. “These initiatives will lead to more convenient and secure experiences across multiple devices by utilizing Samsung’s AI-driven software, platform innovations, and security technologies.”

    Moreover, the company disclosed plans to unify the software experience across its primary product lines—including mobile devices, televisions, and home appliances—under the banner of One UI in the following year. By ensuring a unified product experience and committing to software updates for as long as seven years, Samsung aims to drive continued innovation for its customers.

    To fulfill its vision of AI for All, Samsung will also focus on creating AI solutions for good, enabling innovators to harness new technologies to tackle societal challenges and build a better future for the next generations.

    Tags: , ,