Workflow
RISC-V+AI
icon
Search documents
中国开放指令生态联盟秘书长包云岗:RISC-V产业落地仍需解决四个问题
Core Insights - The RISC-V ecosystem in China is projected to dominate the global market by 2030, with a market size expected to reach $9.27 billion and a compound annual growth rate (CAGR) of 47.4% from SHD Group's forecast [1] - By 2027, RISC-V cores from China are anticipated to account for 50% of global shipments, according to data from the RISC-V Foundation [1] Industry Challenges - The transition to RISC-V faces several challenges, including the existing reliance on ARM architectures, the unclear application scenarios for RISC-V chips, and the shortage of professionals knowledgeable in RISC-V technology [1] - Four critical issues need to be addressed for the RISC-V industry to thrive: the need for competitive RISC-V products and solutions, a more comprehensive software and hardware toolbox, a sufficient talent pool across various levels, and the establishment of benchmark cases to build industry confidence [1] Cost Reduction and New Business Models - RISC-V's open-source nature allows for structural cost reductions and the creation of new business models, such as a Linux+RedHat model in the chip sector, which can support the internet industry [2] - For instance, developing a 64-core server chip with a production volume of 100,000 units incurs a development cost of approximately 750 million RMB, with IP licensing fees and royalties accounting for about 250 million RMB, which is one-third of the total cost. Utilizing an open-source collaborative development model can save companies 250 million RMB in R&D expenses, representing a 33% reduction [2] Future Trends - The proportion of open-source IP in future chips is expected to increase significantly, with predictions indicating a breakthrough to 0% in the near future [3] - RISC-V's customizability enables optimal hardware-software collaboration and agile development, allowing application development teams to quickly tailor chips based on specific use cases [3] - The demand for AI inference capabilities is evolving, and the combination of RISC-V and AI is anticipated to become a new trend, particularly through close collaboration with CPU designs to meet diverse requirements [3]