Mythic Licenses Codasip's L30 RISC-V Core for Next-Generation AI Processor
Munich, Germany – August 10th, 2021 – Codasip, the leading supplier of customizable RISC-V® embedded processor IP, announced today that Mythic, the pioneering AI processor company with breakthrough analog compute-in-memory technology, has selected Codasip’s L30 (originally Bk3) RISC-V-based core for Mythic’s next generation Analog Matrix Processor (Mythic AMP™).
Mythic uses Codasip RISC-V cores in its revolutionary new M1076 Mythic AMP™, which delivers best-in-class performance, scalability, and power efficiency. Mythic’s Analog Matrix Processor is designed with an array of tiles, and each contains: a large Analog Compute Engine (ACE) to store neural network weights and perform matrix multiplications; local SRAM memory for data being passed between the neural network nodes; a single-instruction multiple-data (SIMD) unit for processing operations not handled by the ACE; and a Codasip processor for controlling the sequencing and operation of the tile. Using Codasip Studio, Mythic customized Codasip’s processor IP to meet the unique computing requirements of the M1076 AMP.
“Codasip’s RISC-V cores and customization toolset helped us accelerate the design of our new M1076 AMP while giving us the flexibility we needed to offer an unparalleled combination of performance and power efficiency,” said Ty Garibay, Vice President of Engineering at Mythic. “We look forward to continuing our engagement with Codasip on the next generation of our groundbreaking processors for cutting-edge AI applications.”
“We have been thrilled to see the launch of Mythic’s revolutionary Analog Matrix Processor,” said Dr. Karel Masarík, CEO and Founder of Codasip, “We are very excited to have been chosen for the company’s next artificial intelligence processor design.”
The Codasip L30 processor is based on the RISC-V open instruction set architecture (ISA). It is optimized for low power and area efficiency. It has a single 3-stage processor pipeline architecture, optional caches, optional Floating Point Unit, Multiplication and Division, JTAG and RISC-V debug, and industry standard bus interfaces (AMBA). It also includes support for privilege-mode and memory protection via standard RISC-V PMP, making it an attractive alternative to legacy, proprietary microcontroller cores. L30 is fully configurable and extensible in compliance with the RISC-V standard.
About Codasip
Codasip delivers leading-edge RISC-V processor IP and high-level processor design tools, providing IC designers with all the advantages of the RISC-V open ISA, along with the unique ability to customize the processor IP. As a founding member of the RISC-V Foundation and a long-term supplier of LLVM and GNU-based processor solutions, Codasip is committed to open standards for embedded processors. Formed in 2014 and headquartered in Munich, Germany, Codasip currently has offices in Europe and China, with sales representatives worldwide. For more information about our products and services, visit www.codasip.com. To learn more about RISC-V, visit www.riscv.org.
|
Codasip Hot IP
Related News
- Fractile Licenses Andes Technology's RISC-V Vector Processor as It Builds Radical New Chip to Accelerate AI Inference
- Sophgo Licenses SiFive RISC-V Processor Cores to Drive High-Performance AI Computing Innovation
- Codasip announces next-generation RISC-V processor family for Custom Compute
- Vidtoo Technology Licenses Codasip's Bk3 RISC-V Processor for High-Performance Computing SoC
- GreenWaves Technologies Licenses Intrinsic ID Hardware Root of Trust for RISC-V AI Application Processor
Breaking News
- Ubitium Debuts First Universal RISC-V Processor to Enable AI at No Additional Cost, as It Raises $3.7M
- TSMC drives A16, 3D process technology
- Frontgrade Gaisler Unveils GR716B, a New Standard in Space-Grade Microcontrollers
- Blueshift Memory launches BlueFive processor, accelerating computation by up to 50 times and saving up to 65% energy
- Eliyan Ports Industry's Highest Performing PHY to Samsung Foundry SF4X Process Node, Achieving up to 40 Gbps Bandwidth at Unprecedented Power Levels with UCIe-Compliant Chiplet Interconnect Technology
Most Popular
- Cadence Unveils Arm-Based System Chiplet
- CXL Fabless Startup Panmnesia Secures Over $60M in Series A Funding, Aiming to Lead the CXL Switch Silicon Chip and CXL IP
- Esperanto Technologies and NEC Cooperate on Initiative to Advance Next Generation RISC-V Chips and Software Solutions for HPC
- Eliyan Ports Industry's Highest Performing PHY to Samsung Foundry SF4X Process Node, Achieving up to 40 Gbps Bandwidth at Unprecedented Power Levels with UCIe-Compliant Chiplet Interconnect Technology
- Arteris Selected by GigaDevice for Development in Next-Generation Automotive SoC With Enhanced FuSa Standards
E-mail This Article | Printer-Friendly Page |