Week In Review: Design, Low Power

Simulating 3D-ICs; IP for chiplet developers; diamond quantum microprocessors.


Tools & IP
Ansys updated its product suite, adding new tools and workflows in Ansys 2022 R1. It adds Phi Plus meshing technology to improve simulation of PCB and complex 3D IC packaging. It also introduces RedHawk-SC SigmaDVD, a statistically realistic modeling technique to identify the worst-case dynamic voltage-drop in hours and make it possible to achieve near 100% coverage of all relevant switching scenarios of neighboring cells. Additionally, Ansys Sherlock features a new, semi-automated workflow that takes advantage of integrations with Ansys AEDT Icepak to deliver more predictively accurate thermal analysis simulations for PCBs.

Additionally, Ansys simulation tools are now available on the cloud through Ansys Gateway for Amazon Web Services (AWS). Gateway allows customers to access existing Ansys software applications and will be able to create, customize, and connect Ansys software with 3rd party applications using minimal technical skills via a user-friendly interface. “Cloud computing accelerates and expands simulation by removing hardware barriers and computer restrictions, streamlining all processes and workflows,” said Shane Emswiler, Senior Vice President of Products at Ansys.

eTopus Technology, QuickLogic, and OpenFive teamed up on a platform of interoperable base IP that can be integrated by chiplet developers with minimal risk and reduced development costs. Initially, two process nodes are being supported including 22nm FDX and 7/6nm covering high-value and high-performance markets. The offering includes SerDes IP for 1-56G LR and for PCIe 5/6, eFPGA IP, Die2Die controller IP, Ethernet MAC IP, and RISC-V IP.

isMedia integrated Mixel’s MIPI C-PHY/D-PHY IC into its UCI72S frame grabber product and plans to use it in its next generation frame grabber, UCI73S. The isMedia UCI72S and UCI73S frame grabbers are designed to support various sensor applications and supports multiple video formats including: 8bit to 16bit Bayer (8/10/12/14/16), 8bit or 16bit YUV4:2:2, and RGB888/RGB565 (MIPI CSI2). The UCI72S can support a raw video resolution of up to 16382 x 16382-pixel resolution with built-in 16Gb DDR4 RAM, while the next generation model UCI73S can support a raw video resolution of up to 32764 x 16382-pixel resolution with built-in 32Gb DDR4 RAM. “We have known about Mixel and their MIPI expertise for many years so when we needed an out-of-the box solution for our latest MIPI product, we approached them for help,” said Hyungyu Lee, Head of R&D at isMedia. “We worked closely with Mixel’s team to develop a robust solution using the Mixel IC for our customers that would have otherwise not been possible.”

Arasan Chip Systems launched MIPI Display IP with MIPI DSI-2 and integrated VESA DSC IP. The IP has low gate count, low latency, and reduced memory size implementations and provides real-time compression of high-definition streams up to 8K in resolution.

Altice Labs selected Keysight’s O-RAN Radio Architect (KORA) solutions to advance development of O-RAN equipment and deliver certification services. “We’re pleased to provide Altice Labs with O-RAN test solutions for efficient deployment of 4G and 5G networks based on open, disaggregated architectures,” said Scott Bryden, vice president and general manager at Keysight. “The KORA portfolio consists of integrated solutions based on common software platforms, enabling Altice to validate interoperability, performance, security and compliance of O-RUs to standards specified by the O-RAN Alliance.”‘

LeapMind announced an AI image processing model which operates in real time on edge devices. It uses a low bit quantization technology and the company’s ultra-low power consumption AI inference accelerator IP, Efficiera, to reduce noise and improve the image quality of noisy images.

Infineon reported first quarter 2022 financial results with revenue of €3.16 billion (~$3.61B), up 20% from the same period last year. The company raised its outlook and now expects revenue of about €13.0 billion (~$14.87B) for the year. “Infineon has made a successful start into the 2022 fiscal year. We were able to significantly increase both our revenue and our Segment Result further,” said Reinhard Ploss, CEO of Infineon. “Demand for our products and solutions remains very strong. Utilization of our manufacturing capacities is very high and we are expanding them step by step. This will help us improve the availability of products that we manufacture in-house over the course of the year. Overall, demand for semiconductors is outstripping supply by far. Electrification and digitalization continue to drive substantial growth in our target markets. We expect the supply situation in some application areas to remain tight well into the current calendar year.”

Quantum & HPC
Quantum hardware manufacturer Quantum Brilliance, the Fraunhofer Institute for Applied Solid State Physics IAF, and the University of Ulm are beginning a joint research project on diamond-based quantum microprocessors. The project, funded with €19.9 million (~$22.5 million), aims to develop atomically precise techniques for the fabrication of quantum microprocessors and to find new methods for selective initialization, read-out, and manipulation of qubits in quantum computers with multiple processor nodes.

The Government of Quebec and IBM are teaming up to launch a technology hub to develop new projects, collaborations, and skills-building initiatives in areas of research such as energy, life sciences, and sustainability. The Quebec-IBM Discovery Accelerator will deploy an IBM Quantum System One, the company’s first in Canada, and leverage a base of high-performance computing (HPC) in order to integrate classical and quantum technology to explore various scientific challenges.

The Indian Institute of Science (IISc) has installed and commissioned a new supercomputer. With a supercomputing capacity of 3.3 petaflops, Param Pravega is one of the most powerful supercomputers in the country, and the largest in an Indian academic institution. Designed by the Centre for Development of Advanced Computing (C-DAC), it is a mix of heterogeneous nodes, with Intel Xeon Cascade Lake processors for the CPU nodes and NVIDIA Tesla V100 cards on the GPU nodes. The hardware consists of an ATOS BullSequana XH2000 series system. It will be used in a wide range of research areas, including genomics and climate research.

Leave a Reply

(Note: This name will be displayed publicly)