日前，一则重磅消息惊动全球半导体产业圈。 当地时间12月10日晚，全球半导体巨头高通公司宣布完成对 Ventana Micro Systems的收购，进一步强化 其在RISC-V领域的技术布局。 巨头加码RISC-V，印证了AI时代RISC-V开源架构的广阔前景。而在此背后，一座中国城市的战略布局 也浮出水面。 去年9月，Ventana携手珠海科技产业集团、跃昉科技在珠海牵头成立了全球首个基于RISC-V的DSA产业 创新合作组织——RDSA产业联盟。业内人士分析，随着Ventana正式纳入高通体系，该联盟有望接入高 通覆盖全球的技术、渠道与生态资源，迎来国际化发展的重要契机。 早在2023年12月召开的中共珠海市委九届六次全会上，珠海市委书记、省委横琴工委书记陈勇就提出 以"云上智城"、低空经济、开源体系等为重点大力培育新质生产力。2024年9月，陈勇书记进一步提 出，要借助RDSA产业联盟等平台，汇集全球产业链智慧和力量，将珠海打造为具有重要国际影响力和 示范性的RISC-V创新高地。 2025年RISC-V产业发展大会暨RDSA国际论坛在珠海、澳门两地举行 从前瞻布局"RISC-V+芯粒+DSA"高壁垒 ...

RT DHH (@dhh)Europe can't fine or regulate its way back to technological relevance. Monopoly interventions must be based on simple economics. Having Brussels design the color of the checkmark is retarded. The DSA, DMA, and even GPDR has got to go. Full reboot required. ...

RT Martin D (@SkyMartiner)Mike Benz is spot on. All force needed to render the coming additional EU "fines" of VLOPS in the US (X, Facebook) useless and destructive.The DSA is largely a grand theft scheme to rake in $$$ by innovative losers in the EU. Should be #1 priority in the State Department. ...

Core Insights - RISC-V is predicted to capture 25% of the semiconductor market by 2030, with chip shipments reaching 17 billion units, showcasing remarkable growth for an architecture that has been in development for just over a decade [1] - Sunzhang Technology is emerging as a significant player in the RISC-V space, offering unique services that combine IP and EDA to provide comprehensive processor solutions [1][4] RISC-V and Market Trends - The chip industry is evolving, with a distinction between general-purpose and domain-specific chips. The limitations of general-purpose architectures are becoming apparent as industry demands diversify [3] - Sunzhang Technology focuses on refining RISC-V IP to create a robust product family that meets diverse application needs, recognizing the inefficiencies of general architectures in specialized scenarios [3][4] DSA Methodology - The DSA (Domain Specific Architecture) approach is built on validated RISC-V IP, allowing for deep optimization tailored to specific client needs, enhancing processor performance for targeted applications [4] - Sunzhang Technology has developed tools that enable clients to design processors suited to their unique scenarios, facilitating the implementation of the DSA methodology [4][6] Product Offerings - Sunzhang Technology has established a comprehensive RISC-V IP product line, including M050, M100, M130, M500, and M510 series, which have been commercially validated and outperform competitors in benchmarks [6] - The products are utilized across various sectors, including communications, automotive electronics, and IoT, with some achieving ISO 26262 ASIL-D certification for automotive reliability [6] EDA Development Platform - The ArchitStudio platform accelerates the design and implementation of DSA processors, allowing developers to focus on architecture and instruction set design rather than complex RTL-level tasks [7] - This platform significantly reduces the development workload and costs associated with processor design, streamlining the process for clients [7] Tailored Solutions - Sunzhang Technology has created solutions like the "Zhiying" series for edge AI devices, addressing the demand for low power, small size, and high performance [10] - The combination of ArchitStudio and customizable NPU modules enables precise matching of AI models to various scenarios, tackling challenges in real-time response, energy consumption, and cost optimization [10] Strategic Vision - In a competitive market, RISC-V DSA is seen as a strategic choice for building technological barriers and helping clients differentiate themselves [12] - Sunzhang Technology aims to unify tools and frameworks to address potential fragmentation issues in compiler and SDK development, aspiring to become a leading provider of processor solutions globally [12]

Core Viewpoint - The development of AI is fundamentally changing the software programming paradigm, leading to the emergence of Software 3.0, where natural language prompts are replacing traditional programming code, and large language models (LLMs) are becoming the new programming interface [2][3]. Group 1: Software Evolution - Software 1.0 was characterized by human-written code, while Software 2.0 shifted to neural networks, requiring data preparation and parameter training [2]. - Software 3.0 represents a significant transformation in software development, driven by the rise of large language models [2]. - The transition to Software 3.0 necessitates advancements in hardware, referred to as Hardware 3.0, to support new computational demands [2][3]. Group 2: Hardware Requirements - The dominance of CPUs in Software 1.0 has shifted to GPUs in Software 2.0 due to the need for parallel processing capabilities [3]. - The rapid development of transformer-based models in Software 3.0 has led to the increased adoption of Domain-Specific Architectures (DSA) [5]. - A balance between specialized efficiency and programming generality is crucial for the development of Hardware 3.0 [5][8]. Group 3: Challenges in AI Processor Design - Key challenges in designing AI processors include the lengthy time required to construct AI computing architectures, the prolonged development of instruction systems, and the long cycles for compiling software [9]. - Achieving widespread ecosystem support for self-built instruction systems presents significant hurdles [9]. Group 4: RISC-V and EVAS Architecture - RISC-V's open and modular design allows for the customization of AI acceleration instruction sets, making it a suitable foundation for DSA [8]. - The introduction of the Virtual Instruction Set Architecture (VISA) aims to bridge the gap between AI compilers and backend compilation, enhancing performance optimization [10][11]. - The EVAS architecture integrates VISA with RISC-V microinstructions, ensuring efficient execution of AI computations while improving user programming experience [12][16]. Group 5: Upcoming Innovations - The upcoming chip from the company will support various data types, including INT4, INT8, FP8, FP16, and BF16, with a focus on mixed-precision computing [17]. - The new architecture aims to provide advanced computing solutions for applications in autonomous driving, embodied intelligence, and other edge-cloud industry applications, contributing to the progress towards AGI [17].