David Harold

2 days ago | jonpeddie.com | David Harold

At the RISC-V EU Summit, Arteris CMO Michal Siwinski emphasized the critical role of Network-on-Chip (NoC) technology in enabling efficient data movement, especially as SoCs grow more complex. He discussed balancing NoC area costs, the importance of soft IP tailored to foundry needs, and NoCs’ impact on heterogeneous systems and memory design. Arteris blends manual and AI-assisted tools, generating 70% of revenue from IP and 30% from software.

3 days ago | jonpeddie.com | David Harold

Finnish start-up Flow Computing has entered alpha testing for its compiler that enables CPUs integrated with its Parallel Processing Unit (PPU) to achieve up to 100x performance gains on parallel workloads. The milestone supports Flow’s plans to license its architecture to CPU vendors across Arm, RISC-V, x86, and Power platforms.Flow Computing, a Finnish start-up aiming to redefine CPU performance scaling, has reached a key milestone on its path to commercialization.

1 week ago | jonpeddie.com | David Harold

Tenstorrent, led by chip legend Jim Keller, is developing next-gen AI processors using open RISC-V CPUs and modular chiplets. Its Athena CPU chiplet and upcoming Callandor processor form the core of a scalable, AI-ready platform. Backed by $693 million in funding from Samsung, LG, Hyundai, and others, the company is licensing its IP globally and partnering in Japan and Korea to co-develop advanced AI chips.

Feb 24, 2025 | jonpeddie.com | David Harold

Andrea Gallo from RVI speaks to JPR’s David Harold about RISC-V’s future. RISC-V is moving towards further standardization with the RVA23 Profile, ensuring software compatibility while retaining flexibility. AI and HPC are key focus areas, with new matrix extensions optimizing workloads. Gallo sees hardware-software co-design as the future, reinforcing RISC-V’s customization advantage over legacy architectures.

Jan 29, 2025 | jonpeddie.com | David Harold

Qualcomm announced major automotive deals at CES 2025, with Leapmotor, Mahindra, and Hyundai Mobis adopting its Snapdragon Digital Chassis. Its hardware is proven, but software remains the challenge. Rivals like Mobileye and Nvidia lead in ADAS, and automakers demand reliability. Qualcomm’s Snapdragon Ride Vision Systemaims to compete, but trust takes time. A partnership with Amazon highlights its AI ambitions, but execution is key.