Samsung Electronics grew net profit by more than 400 percent in Q1, to $4.91 billion, on revenue of about $52.3 billion, a nearly 13 percent increase year-over-year. The results were credited mainly to higher memory chip prices resulting from AI demand buoying the company’s semiconductor business. Solid performance in smartphones — with the launch of its Galaxy S24 series, the first to pack AI-optimized chips — supported the stellar performance. It was a dramatic rebound from 2023, when post-COVID economic fallout drove Samsung to a 15-year profit low and semiconductor losses of almost $11 billion. Continue reading Samsung Chip Rebound Sends Q1 Net Profit Up 400 Percent
By
Paula ParisiFebruary 1, 2024
Canon is gearing up to begin shipping its new nanoimprint lithography chipmaking machines, possibly this year. The equipment — which uses a stamping process Canon says will be cheaper and more energy-efficient than ASML’s light-based extreme ultraviolet (EUV) technology — could be a disruptor in a market dominated by the Dutch company. Such machines, essential in manufacturing semiconductors, imprint circuitry onto silicon wafers in patterns that can be thousands of times thinner than the width of a human hair. Foundries TSMC, Samsung Semiconductor and Intel rely on ASML’s EUV machines in the manufacture of high-end chips. Continue reading Canon Could Begin Shipping Chip Stamping Machine in 2024
By
Paula ParisiAugust 2, 2022
Nvidia has issued a software update for its formidable NeMo Megatron giant language training model, increasing efficiency and speed. Barely a year since Nvidia unveiled Megatron, this latest improvement further leverages the transformer engine architecture that has become synonymous with deep learning since Google introduced the concept in 2017. New features result in what Nvidia says is a 5x reduction in memory requirements and up to a 30 percent gain in speed for models as large as 1 trillion parameters, making NeMo Megatron better at handling transformer tasks across the entire stack. Continue reading Nvidia Turbo Charges NeMo Megatron Large Training Model
By
Debra KaufmanFebruary 5, 2020
Over 50 million people worldwide use Apple’s virtual assistant Siri. Apple, focused on improving Siri’s capabilities, published research on how to improve voice trigger detection, speaker verification and language identification for multiple speakers. Apple researchers suggest that an AI model be trained for automatic speech recognition and speaker recognition. Rather than approach it as two independent tasks, the researchers proved that those tasks might actually help one another to “estimate both properties.” Continue reading Apple Researchers Improving Accuracy of Virtual Assistant
By
Rob ScottDecember 20, 2019
Rather than splitting itself into separate product and IP businesses as it explored earlier this year, Rovi-owned TiVo is merging with Xperi Corporation in a deal valued at $3 billion. San Jose-based TiVo, a pioneer in the DVR industry, licenses its IP for areas involving digital rights management, electronic program guides and metadata. Xperi, also with headquarters in San Jose, is a tech licensor in areas including mobile computing, data and memory storage and 3D integrated circuits. The deal will integrate TiVo and Xperi’s IP licensing and product businesses, which will then operate as separate units so that one could be sold in the future. Continue reading $3 Billion TiVo-Xperi Merger Is All About the Patent Portfolio
By
Debra KaufmanAugust 26, 2019
Los Altos, CA-based startup Cerebras, dedicated to advancing deep learning, has created a computer chip almost nine inches (22 centimeters) on each side — huge by the standards of today’s chips, which are typically the size of postage stamps or smaller. The company plans to offer this chip to tech companies to help them improve artificial intelligence at a faster clip. The Cerebras Wafer-Scale Engine (WSE), which took three years to develop, has impressive stats: 1.2 trillion transistors, 46,225 square millimeters, 18 gigabytes of on-chip memory and 400,000 processing cores. Continue reading Cerebras Builds Enormous Chip to Advance Deep Learning
By
Debra KaufmanJuly 17, 2018
Intel is acquiring eASIC, a 120-person custom chip company in Silicon Valley, to help boost its Programmable Solutions Group. The company stated that, “FPGAs [field programmable gate arrays] are experiencing expanding adoption due to their versatility and real-time performance,” and that “eASIC has a proven, 19-year success record … [and its] addition … will help us meet customers’ diverse needs of time-to-market, features, performance, cost, power and product life cycles.” Terms of the deal were not revealed. Continue reading Intel to Purchase eASIC to Expand Programmable Solutions
By
Debra KaufmanJuly 17, 2018
Apple’s latest 15-inch and 13-inch MacBook Pros feature improvements in processor speeds, RAM and internal storage, and allow the user to access Siri with voice commands. The new MacBook Pros also feature a much quieter keyboard. The 15-inch version, with a six-core processor, is estimated to be 70 percent faster than the previous model, and the amount of memory and storage is double the amount of the previous model. It also gave a seal of approval to Blackmagic’s eGPU peripheral, ideal for intensive graphics use. Continue reading Apple Targets Creatives With MacBook Pro Speed, Features
By
Rob ScottJanuary 15, 2018
During his keynote presentation at CES 2018 in Las Vegas, Qualcomm president Cristiano Amon discussed the future of 5G, predicting that flagship 5G smartphones will hit shelves in 2019 and overtake the market quickly. The rollout “will occur ‘very fast’ because of 5G’s 30x improvement in latency, which lets phone makers keep the price of phones down by offloading phone memory to the cloud,” reports TWICE. The presentation suggested that “the 5G transition is going to be faster than the 4G transition, and it will create a ‘massive’ Internet of Things, enabling new business models and making autonomous vehicles a reality.” Continue reading Qualcomm CES Keynote: 5G Transition to Launch Massive IoT
By
Debra KaufmanAugust 1, 2017
Samsung Electronics’ semiconductor unit posted Q2 sales of $15.7 billion and operating profit of $7.1 billion, surpassing Intel’s Q2 earnings of $14.8 billion and operating profit of $3.8 billion. This marks the first time in nearly 25 years that Intel has lost its No. 1 position in the chip-making market. Up until this quarter, Santa Clara, CA-based Intel has enjoyed a comfortable lead over Samsung’s chip unit, which got its start with a focus on mobile devices. The semiconductor industry is currently valued at $365 billion. Continue reading Intel Is Topped by Samsung After 25-Year Lead in Chip Sales
By
Debra KaufmanMay 18, 2017
Hewlett Packard Enterprise introduced a prototype computer dubbed The Machine, which has a single 160-terabyte memory bank that can simultaneously process the data found in 160 million books. This never-before-accomplished feat reveals the potential of what’s called Memory-Driven Computing. The custom-built computer is, says HPE, the world’s largest single-memory computer, based on the largest R&D program in HPE’s history. The Machine, it adds, and its capabilities will be transformational. Continue reading HP Introduces Memory-Driven Computing With ‘The Machine’
By
Debra KaufmanApril 14, 2017
Advanced Micro Devices (AMD) has acquired Nitero, a startup responsible for a 60-gigahertz wireless chip that transmits high-res video without latency. AMD, which bought the company for an undisclosed price, believes that Nitero’s chip will enable it to push sales of more wireless virtual reality headsets. Sales of VR headsets, according to AMD executive Roy Taylor, have been limited due to their need to be tethered to a computer. Nitero was originally a spinoff from a research center sponsored by the Australian government. Continue reading AMD Pitches Latency-Free Virtual Reality via Super-Fast Wi-Fi
By
Debra KaufmanMarch 21, 2017
Intel is shipping its first products based on 3D XPoint, a technology it has been developing for more than ten years to bridge the gap between conventional memory, which holds data for immediate use, and longer-term storage. The company reports that 3D XPoint is 1,000 times faster than NAND flash often used in storage drives, but only 1 percent the speed of DRAM (dynamic random-access memory) temporary memory. Intel says it can speed up tasks in fraud detection, retail recommendations and autonomous vehicles. Continue reading Intel Debuts Memory Products Based on 3D XPoint Technology
By
Debra KaufmanFebruary 13, 2017
Apple may dub its next smartphone either iPhone 8 or iPhone X, but whatever the name, it will be a special 10th-anniversary version and, says a knowledgeable source, is likely to cost more than $1,000. That’s not really a stretch when you note that the 256GB version of the iPhone 7 Plus already sells for $969. The iPhone 8 will feature a range of new features, including an OLED display and a memory upgrade, both of which are expensive. The 5.8-inch phone is expected to look like a smooth black monolith. Continue reading Apple’s iPhone 8 to Feature Wireless Charging, OLED Screen
By
ETCentricJanuary 25, 2017
Netflix is accommodating Android users who are interested in offline downloads but face storage limitations with their mobile devices. The streaming video service now allows users to store TV shows and movies on microSD cards. “The latest version of the Netflix app gives you a choice of saving things to internal storage or, if your phone supports SD, picking the external option instead,” reports The Verge. Previously, customers could only save to internal memory, which frustrated those “with 32GB of built-in space and a ton of expandable storage.” The offline Netflix content has time limits, however, and eventually expires. Continue reading Android Users Can Download Netflix Content onto SD Storage