Week In Review: Design, Low Power


MLCommons debuted the latest results for the MLPerf Inference v3.0 and Mobile v3.0 benchmark suites, which measure the performance and power-efficiency of applying a trained machine learning model to new data in data center, edge, and mobile use cases. Overall, MLCommons said the results showed both power efficiency improvements and significant gains in performance in some benchmark tests. Seve... » read more

Week In Review: Design, Low Power


Tools, IP, chips Synopsys unveiled a new data-visibility and machine intelligence-guided design optimization solution. DesignDash is complementary to the company's DSO.ai AI-driven design-space-optimization tool and provides a real-time, unified, 360-degree view of all design activities. It uses deep analytics and machine learning to extract and reveal actionable understanding from large amoun... » read more

Week In Review: Manufacturing, Test


GlobalFoundries launched GF Labs, an “open framework of internal and external research and development initiatives that deliver a differentiated pipeline of market-driven process technology solutions for future data-centric, connected, intelligent and secure applications.” Greg Bartlett, GF's senior vice president of technology, engineering at quality, said the goal is to develop and exp... » read more

Week In Review: Manufacturing, Test


Deals AMD plans to purchase cloud startup Pensando for about US $1.9 billion. In a presentation at the SEMI ISS conference this week, AMD CTO Mark Papermaster described Pensando's technology as a "highly programmable packet-processing engine that allows you to speed up systems designed for the data center." Intel, Micron, Analog Devices and MITRE Engenuity formed an alliance to accelerate c... » read more

Week In Review: Manufacturing, Test


The U.S. Senate approved the 2022 America COMPETES act, which has big ramifications for the chip industry. The bill now heads to the House for further reconciliation. If approved, it would provide more than $50 billion in U.S. subsidies for semiconductor chip manufacturing. The SIAC (Semiconductor In America Coalition) urged Congress to act promptly to achieve a bipartisan compromise soon and o... » read more

Week In Review: Design, Low Power


Tools Synopsys introduced a new model for using its EDA tools on the cloud. Synopsys Cloud provides pay-as-you-go access to the company's cloud-optimized design and verification products, with pre-optimized infrastructure on Microsoft Azure to address higher levels of interdependencies in chip development. "As more design flows incorporate AI, requiring even more resources, the virtually unlim... » read more

Week In Review: Design, Low Power


Intellectual Property Flex Logix inked an agreement with the Air Force Research Laboratory, Sensors Directorate (AFRL/RY) covering any Flex Logix IP technology for use in all US Government-funded programs for research and prototyping purposes with no license fees. “Our first license with AFRL for EFLX eFPGA in GlobalFoundries 12nm process was highly successful, with more than a half dozen pr... » read more

Week In Review: Manufacturing, Test


Packaging ASE, AMD, Arm, Google, Intel, Meta, Microsoft, Qualcomm, Samsung, and TSMC have announced the formation of a consortium that will establish a die-to-die interconnect standard and foster an open chiplet ecosystem. The founding companies also ratified the UCIe specification, an open industry standard developed to establish a standard interconnect at the package level. The UCIe 1.0 s... » read more

Week In Review: Manufacturing, Test


Chipmakers, OEMs UMC plans to build a new fab next to its existing 300mm fab in Singapore. The new fab, called Fab12i P3, will manufacture wafers based on UMC’s 22nm/28nm processes. The planned investment for this project will be $5 billion. The first phase of this greenfield fab will have a monthly capacity of 30,000 wafers with production expected to commence in late 2024. To account fo... » read more

Week In Review: Design, Low Power


Tools & IP Codasip debuted two new customizable low power embedded RISC-V processor cores. To support embedded AI applications, the L31/L11 cores run Google’s TensorFlowLite for Microcontrollers. Codasip Studio tools can be used to customize for specific system, software, and application requirements. Licensing the CodAL description of a Codasip RISC-V core grants customers a full archit... » read more

← Older posts Newer posts →