Open side-bar Menu
 EDACafe Editorial
Roberto Frazzoli
Roberto Frazzoli
Roberto Frazzoli is a contributing editor to EDACafe. His interests as a technology journalist focus on the semiconductor ecosystem in all its aspects. Roberto started covering electronics in 1987. His weekly contribution to EDACafe started in early 2019.

Arm goes public; Nvidia record results; TSMC’s European joint venture

 
September 1st, 2023 by Roberto Frazzoli

Catching up on some of the news from the last thirty days or so, let’s start with the upcoming change of Synopsys’ top management: on January 1st 2024, Sassine Ghazi will replace Aart de Geus as Synopsys’ Chief Executive Officer. Ghazi assumed the role of Synopsys COO in August 2020 and was appointed to the role of president in November 2021. De Geus (69), founded Synopsys in 1986.

Arm to go public

As previously announced, Arm is going public. On August 20, the company announced that it has publicly filed a registration statement with the U.S. Securities and Exchange Commission relating to its initial public offering on the Nasdaq Global Select Market under the symbol “ARM”.

Nvidia Q2 record results

Nvidia reported record results for the second quarter ended July 30, 2023: global revenue was $13.51 billion, up 88% from Q1 and up 101% from year ago; Data Center revenue was $10.32 billion, up 141% from Q1 and up 171% from year ago. Nvidia has also announced an expanded partnership with Google Cloud which will include the general availability of purpose-built Google Cloud A3 virtual machines powered by Nvidia H100 GPUs.

Fab and foundry updates

Quickly recapping some late July news, Analog Devices is investing more than $1 billion to expand its semiconductor wafer fab in Beaverton, Oregon; and Silicon Box, a Singapore-based startup, has launched its USD 2 billion packaging facility specializing in chiplet solutions using a proprietary interconnection technology.

Moving to more recent updates, Infineon will significantly expand its Kulim, Malaysia, fab to build the world’s largest 200-millimeter silicon carbide power fab. Over the next five years the company will additionally invest up to five billion euros in Kulim during a second construction phase.

TSMC, Bosch, Infineon and NXP have announced a plan to jointly invest in a new company (European Semiconductor Manufacturing Company, ESMC) which will be based in Dresden, Germany, with the goal of building a 300mm fab – operated by TSMC – to support the future capacity needs of the automotive and industrial sectors. The 10 billion euro project is expected to be supported by the European Chips Act. Production start is scheduled for end of 2027, with a monthly capacity of 40,000 wafers on TSMC’s 28/22 nanometer planar CMOS and 16/12 nanometer FinFET process technology. The joint venture will be 70% owned by TSMC, with Bosch, Infineon, and NXP each holding 10% equity stake.

Tower Semiconductor has mutually agreed with Intel to terminate their previously announced merger agreement. As stated in a press release, a reason for the decision was the lack of indications regarding certain required regulatory approval.

Samsung Electronics has reportedly established a research and development organization named “Samsung Federal Inc.” in Silicon Valley with the aim of enhancing its competitiveness in next-generation semiconductors. This decision is seen as a response to the growing uncertainties caused by economic downturns and trade tensions between the United States and China.

Euro chipmakers and Qualcomm to invest on Risc-V in automotive

Bosch, Infineon, Nordic, NXP, and Qualcomm will jointly invest in a company aimed at advancing the adoption of Risc-V globally by enabling next-generation hardware development. Formed in Germany, this company will aim to accelerate the commercialization of future products based on the Risc-V architecture. Initial application focus will be automotive, with a possible expansion to include mobile and IoT.

Tachyum improves its chip after switching to new EDA tools

AI processor startup Tachyum has announced that new EDA tools, utilized during the physical design phase of its Prodigy Universal Processor, have allowed the company to achieve significantly better results than previously anticipated – including an increase in the number of Prodigy cores to 192. As explained in a press release, the replacement of RTL simulation and physical design tools became necessary after the Prodigy design team had to replace some outsourced IP blocks. The story was widely reported by the media one year ago; see, for example, these 2022 articles by eeNews Europe and The Register. Now, with a new set of EDA tools from an unnamed EDA vendor, Tachyum was able to increase the number of cores by 50 percent, and SERDES from 64 to 96 on each chip. According to the company, die size grew minimally, from 500mm2 to 600mm2. Tachyum claims to have achieved other improvements during the physical design stage.

Free Gerber viewer aims to bridge the gap between PCB design and manufacturing

Chinese PCB manufacturer NextPCB has released its HQDFM Gerber viewer – a free, online tool for displaying and reviewing PCB Gerber files and OBD++ files, which also incorporates the company’s Design for Manufacture algorithms. According to the company, HQDFM seeks to resolve the problems caused by the division between PCB design and PCB manufacture. The tool allows designers to navigate and analyze the production files and check for design issues that may impact manufacturing or cause long-term reliability issues.

UCIe 1.1

The UCIe Consortium has recently released the UCIe 1.1 specification. The new release extends reliability mechanisms to more protocols and supports broader usage models. Additional enhancements are included for automotive usages – such as predictive failure analysis and health monitoring – and enabling lower-cost packaging implementations. The specification also details architectural specification attributes to define system setups and registers that will be used in test plans and compliance testing to ensure device interoperability. The UCIe 1.1 specification is fully backward compatible with the UCIe 1.0 specification.

Acquisitions

Sensor and IC vendor Allegro MicroSystems has signed a definitive agreement to acquire Crocus Technology for $420 million in cash. Crocus specializes in advanced Tunnel Magnetoresistance (TMR) sensor technology.

AMD has acquired France-based Mipsology, a developer of AI inference software.

Macom, a supplier of semiconductor products, has entered into a definitive agreement to acquire the radio frequency business of Wolfspeed, which includes a portfolio of Gallium Nitride on Silicon Carbide products used in high performance RF and microwave applications.

Norway-based Nordic Semiconductor has entered into an agreement to acquire the IP portfolio of Atlazo, a US-based artificial intelligence and machine learning company.

Renesas will acquire France-headquartered Sequans through tender offer. Sequans specializes in cellular IoT technology.

Synopsys has completed the acquisition of Germany-based PikeTec, a developer of solutions for the testing and verification of automotive software for control unit systems.

Further reading

Below is part of the transcript from a video, a June episode of the Fully Charged Show podcast where Ford Motor Company CEO Jim Farley candidly talked about the software hurdles legacy carmakers are facing. “If I explain to the listeners how crazy our software system is and why it’s so difficult for legacy car companies to get software right…” he said. “We’ve farmed out the software modules that control the vehicles to our suppliers because we could bid them against each other… The problem is the software is all written by 150 different companies and they don’t talk to each other… we can’t even understand it all,” he continued. “So that’s why at Ford we decided in the second generation product to completely in-source the electrical architecture. To do that, you need to write all the software yourself, but just remember, car companies haven’t written software like this ever… We’re literally writing the software to operate the vehicle for the first time,” Farley pointed out.

This article reports about a theory developed by a group of MIT researchers, envisioning a biologically plausible model of a “transformer” deep learning model. The biological version combines neurons with another type of brain cells called astrocytes.

Logged in as . Log out »




© 2024 Internet Business Systems, Inc.
670 Aberdeen Way, Milpitas, CA 95035
+1 (408) 882-6554 — Contact Us, or visit our other sites:
TechJobsCafe - Technical Jobs and Resumes EDACafe - Electronic Design Automation GISCafe - Geographical Information Services  MCADCafe - Mechanical Design and Engineering ShareCG - Share Computer Graphic (CG) Animation, 3D Art and 3D Models
  Privacy PolicyAdvertise