News Posts matching #Silicon

Return to Keyword Browsing

Apple Silicon Will Support Thunderbolt 4

Since Apple has announced its transition from Intel to its custom Apple Silicon processors, there has been quite a lot of speculations on what the new processors will bring. Just a few days ago, Intel announced the latest advancement of its Thunderbolt port in the form of Thunderbolt 4, which further advances the Thunderbolt standard with a heap of new features. Since Apple has decided to do away from Intel silicon, there has been a question whatever Apple will offer Thunderbolt 4 support on its Macs. And it seems like we don't have to wonder any further. In the statement below, which Apple spokesman gave to The Verge, the answer is loud and clear.
Apple spokesman for The Verge
Over a decade ago, Apple partnered with Intel to design and develop Thunderbolt, and today our customers enjoy the speed and flexibility it brings to every Mac. We remain committed to the future of Thunderbolt and will support it in Macs with Apple silicon.

Google Teams up with SkyWater to Create Open Source PDK for Silicon Design

Silicon design is a hard process. You start by defining design in a hardware description language (HDL) and finish by routing all the wires on the piece of silicon. This used to be done by using proprietary tools and silicon Process Design Kits (PDK) that were unique to silicon manufacturing facilities. Starting from Intel, Samsung, and TSMC, they all have a specific PDK that is used for their silicon manufacturing and it contains all of the specifications for their nodes. It contains design constrains/information on the number of metal layers, how close can two wires be to each other, etc.

In the aim to open-source silicon design and finally allow for fully open silicon design flow, Google has partnered with SkyWater to deliver an open-source PDK that will allow designers to produce silicon on an open platform. Despite being designed for an older 130 nm node, this represents a massive achievement as there was a lack of such a thing before. There is already existing open-source toolchain to create designs, but there wasn't a PDK to port them to. Now, the PDK is available and designers can manufacture 100% open source silicon at SkyWater's facilities. In the GitHub repository listed as a source below, you can find some examples on how you can use the PDK with open source tools as well.

Basemark GPU will be the first benchmark for Apple Silicon based Macs

On Monday 22nd of June Apple announced Mac transition to Apple Silicon. Even though this transition was quite expected, the industry got very excited upon the announcement. Apple released quite a lot of information about their plans, but one key question remained unanswered: how fast are Apple's new ARM based Mac chips?

Apple said people should expect pure performance in one category in particular - graphical performance. What is the performance difference over the Intel integrated graphics that ship in a new MacBook Air? There is no public information available about it.
Basemark Benchmark Apple Silion

Hot Chips 2020 Program Announced

Today the Hot Chips program committee officially announced the August conference line-up, posted to hotchips.org. For this first-ever live-streamed Hot Chips Symposium, the program is better than ever!

In a session on deep learning training for data centers, we have a mix of talks from the internet giant Google showcasing their TPUv2 and TPUv3, and a talk from startup Cerebras on their 2nd gen wafer-scale AI solution, as well as ETH Zurich's 4096-core RISC-V based AI chip. And in deep learning inference, we have talks from several of China's biggest AI infrastructure companies: Baidu, Alibaba, and SenseTime. We also have some new startups that will showcase their interesting solutions—LightMatter talking about its optical computing solution, and TensTorrent giving a first-look at its new architecture for AI.
Hot Chips

Arm Offers Startups Zero-cost Access to its IP Portfolio

Arm today announced the launch of Arm Flexible Access for Startups, an extension of its already highly successful Flexible Access program. This new initiative offers early-stage silicon startups zero-cost access to a huge range of Arm's leading IP, along with global support and training resources, enabling them to start on their journey to commercial silicon and business scale.

"In today's challenging business landscape, enabling innovation is critical - now more than ever, startups with brilliant ideas need the fastest, most trusted route to success and scale," said Dipti Vachani, senior vice president and general manager, Automotive and IoT Line of Business, Arm. "Arm Flexible Access for Startups offers new silicon entrants a faster, more cost-efficient path to working prototypes, resulting in strengthened investor confidence for future funding."
Arm Chip

Huawei Moves 14 nm Silicon Orders from TSMC to SMIC

Huawei's subsidiary, HiSilicon, which designs the processors used in Huawei's smartphones and telecommunications equipment, has reportedly moved its silicon orders from Taiwan Semiconductor Manufacturing Company (TSMC) to Semiconductor Manufacturing International Corporation (SMIC), according to DigiTimes. Why Huawei decided to do is move all of the 14 nm orders from Taiwanese foundry to China's largest silicon manufacturing fab, is to give itself peace of mind if the plan of the US Government goes through to stop TSMC from supplying Huawei. At least for the mid-tier chips built using 14 nm node, Huawei would gain some peace as a Chinese fab is a safer choice given the current political situation.

When it comes to the high-end SoCs built on 7 nm, and 5 nm in the future, it is is still uncertain how will Huawei behave in this situation, meaning that if US cuts off TSMC's supply to Huawei, they will be forced to use SMIC's 7 nm-class N+1 node instead of anything from TSMC. Another option would be Samsung, but it is a question will Huawei put itself in risk to be dependant on another foreign company. The lack of 14 nm orders from Huawei will not be reflecting much on TSMC, because whenever someone decides to cut orders, another company takes up the manufacturing capactiy. For example, when Huawei cut its 5 nm orders, Apple absorbed by ordering more capacity. When Huawei also cut 7 nm orders, AMD and other big customers decided to order more, making the situation feel like there is a real fight for TSMC's capacity.
Silicon Wafer

TSMC on Track to Deliver 3 nm in 2022

TSMC is delivering record results day after day, with a 5 nm manufacturing process starting High Volume Manufacturing (HVM) in Q2 next year, 7 nm process getting plenty of orders and the fact that TSMC just became the biggest company publicly trading in Asia. Continuing with the goal to match or even beat the famous Moore's Law, TSMC is already planning for future 3 nm node manufacturing, promised to start HVM as soon as 2022 arrives, according to JK Wang, TSMC's senior vice president of fab operations. Delivering 3 nm a whole year before originally planned in 2023, TSMC is working hard, with fab construction work doing quite well, judging by all the news that the company is releasing recently.

We can hope to see the first wave of products built using 3 nm manufacturing process sometime around end of year 2022, when the holiday season arrives. Usual customers like Apple and HiSilicon will surely utilize the new node and deliver their smartphones with 3 nm processors inside as soon as the process is ready for HVM.

TSMC Becomes Asia's Most Valuable Company

Taiwan Semiconductor Manufacturing Company, Limited, also known as TSMC shortly became Asia's biggest and the most valuable company with a market cap of over 8.02 trillion New Taiwan Dollars, which roughly translates to 262.75 billion US Dollars. Becoming the biggest Asian company, TSMC's market capitalization has now surpassed Samsung for the first time in the history of company existence.

The underlying reasons for becoming a company that TSMC is today, are plenty. Firstly, they are providing customers with the flexibility of choosing any manufacturing node, whatever it is the latest 7 nm or the older ones like 180 nm. They have a choice whatever they want to use something older and less expensive or something newer for high-performance and lower power. Additionally, TSMC is re-investing a big part of its profits into research and development efforts to stay competitive and deliver only the best technology to its customers, on time.

Panasonic Exits Silicon Manufacturing Business

Panasonic, an electronics manufacturing giant, has today sold its silicon manufacturing business, marking the end of an era of Japanese semiconductor manufacturing. Once a big player in silicon manufacturing scene, particularly in the '80s and '90s era when Japan's silicon output was huge, Panasonic was considered one of the main players in the silicon manufacturing business. However, due to some difficulties like operating a business with a loss of over $215 million yearly, and having to compete with Chinese and Taiwanese silicon manufacturing firms, Panasonic is selling its silicon production lines.

The subsidiary of Panasonic called "Semiconductors Solutions" is being sold to Nuvoton Technology Corporation, a semiconductor company that spun-off from Winbond Electronics Corporation in 2008, where Winbond still owns 61% stake in Nuvoton despite the spinoff. Additionally, Panasonic forecasts a 27% drop in operating profit for this physical year, with the declining semiconductor manufacturing business counted. The reasoning behind this sale is that the company plans to exit all declining businesses that also include LCD manufacturing, as Chinese alternative manufacturers are stiff competition for Panasonic when it comes to pricing and panel output.

Samsung Develops Industry's First 12-Layer 3D-TSV Chip Packaging Technology

Samsung Electronics Co., Ltd., a world leader in advanced semiconductor technology, today announced that it has developed the industry's first 12-layer 3D-TSV (Through Silicon Via) technology. Samsung's new innovation is considered one of the most challenging packaging technologies for mass production of high-performance chips, as it requires pinpoint accuracy to vertically interconnect 12 DRAM chips through a three-dimensional configuration of more than 60,000 TSV holes, each of which is one-twentieth the thickness of a single strand of human hair.

The thickness of the package (720 µm) remains the same as current 8-layer High Bandwidth Memory-2 (HBM2) products, which is a substantial advancement in component design. This will help customers release next-generation, high-capacity products with higher performance capacity without having to change their system configuration designs. In addition, the 3D packaging technology also features a shorter data transmission time between chips than the currently existing wire bonding technology, resulting in significantly faster speed and lower power consumption.

Researchers Build a CPU Without Silicon Using Carbon Nanotubes

It is no secret that silicon manufacturing is an expensive and difficult process which requires big investment and a lot of effort to get right. Take Intel's 10 nm for example. It was originally planned to launch in 2015, but because of technical difficulties, it got delayed for 2019. That shows how silicon scaling is getting more difficult than ever, while costs are rising exponentially. Development of newer nodes is expected to cost billions of Dollars more, just for the research alone and that is not even including the costs for the setting up a manufacturing facility. In order to prepare for the moment when the development of ever-decreasing size nodes becomes financially and physically unfeasible, researchers are exploring new technologies that could replace and possibly possess even better electrical properties than silicon. One such material (actually a structure made from it) is Carbon Nanotube or CNT in short.

Researchers from MIT, in collaboration with scientists from Analog Devices, have successfully built a CPU based on RISC-V architecture entirely using CNTs. Called RV16X Nano, this CPU is currently only capable of executing a classic "Hello World" program. CNT is a natural semiconductor, however, when manufactured, it is being made as a metallic nanotube. That is due to the fact that metallic nanotubes are easier to integrate into the manufacturing ecosystem. Its has numerous challenges in production because CNTs tend to position themselves randomly in XYZ axes. Researchers from MIT and Analog Devices solved this problem by making large enough surfaces so that enough random tubes are positioned well.

Silicon Lottery Starts Selling Binned 3rd Generation AMD Ryzen CPUs

Silicon Lottery, a company specializing in the process called binning which involves testing of CPUs for particular features (overclocking potential in this case), has released its portfolio of 3rd generation of Ryzen CPUs. As of now, they are offering only Ryzen 7 and Ryzen 9 models, covering Ryzen 7 3700X, 3800X and Ryzen 9 3900X. Ryzen 9 3950X is said to be introduced in September and that is the date Silicon Lottery will reveal the information about overclocking potential of that model and frequencies they have achieved. Mid range Ryzen 5 models should be added at later date as well.

ARM Revokes Huawei's Chip IP Licence

As the trade war between the US and China continues to unfold, we are seeing major US companies ban or stop providing service to China's technology giant Huawei. Now, it looks like the trade war has crossed the ocean and reached the UK. This time, UK based ARM Holdings, the provider of mobile chip IP for nearly all smartphones and tablets, has revoked the license it has given Huawei.

According to the BBC, ARM Holdings employees were instructed to suspend all interactions with Huawei, and to send a note informing Huawei that "due to an unfortunate situation, they were not allowed to provide support, deliver technology (whether software, code, or other updates), engage in technical discussions, or otherwise discuss technical matters with Huawei, HiSilicon or any of the other named entities." The news came from an internal ARM document the BBC has obtained.

Intel Again Leader in Silicon Supply Race

Intel was the historic leader in silicon manufacturing and sales from 1993 through 2016, the year it lost its lead to Samsung. The issue wasn't so much to do with Intel, but more to do with market demands at the time - if you'll remember, it was the time of booming DRAM pricing alongside the smartphone demand increase that propagated stiff competition and manufacturers trying to outgun one another in the form of specs. The DRAM demand - and its ridiculous prices, at the time - propelled Samsung towards the top spot in terms of revenue, leaving Intel in the dust.

However, with the decrease in DRAM pricing following the reduce in smartphone demand and increased manufacturing capabilities of semiconductor manufacturers, which flooded the market with product that is being more slowly digested, has led to the drop of the previously-inflated Dram pricing, thus hitting Samsung's revenues enough for Intel to again become "top dog" in the silicon manufacturing world - even as the company struggles with its 10 nm rollout and faced supply issues of their own. As IC Insights puts it, "Intel replaced Samsung as the number one quarterly semiconductor supplier in 4Q18 after losing the lead spot to Samsung in 2Q17. (...) With the collapse of the DRAM and NAND flash markets over the past year, a complete switch has occurred, with Samsung having 23% more total semiconductor sales than Intel in 1Q18 but Intel having 23% more semiconductor sales than Samsung just one year later in 1Q19!".

U.S. Tech Industry, Including Google, Microsoft, Intel, and Qualcomm, Ban Huawei

The United States tech industry has overnight dealt a potentially fatal blow to Chinese electronics giant Huawei, by boycotting the company. The companies are establishing compliance with a recent Executive Order passed by President Donald Trump designed to "stop the import, sale, and use of equipment and services by foreign companies based in countries that are potential adversaries to U.S. interests," particularly information technology security. Google has announced that it will no longer allow Huawei to license Android, and will stop updates and Google Play access to Huawei smartphones. Huawei can still equip its phones with open-source Android, but it cannot use Google's proprietary software, including Google Play Store, Chrome, and all the other Google apps. Intel decided to no longer supply processors and other hardware to Huawei, for use in its laptops and server products. Sales of AMD processors will stop, too. Qualcomm-Broadcom have decided to stop supply of mobile SoCs and network PHYs, respectively. Microsoft decided to stop licensing Huawei to use Windows and Office products.

The ban is a consequence of the U.S. Government placing Huawei on a list of banned entities, forcing all U.S. companies to abandon all trade with it, without prior approval from the Department of Commerce. Trade cuts both ways, and not only are U.S. firms banned from buying from Huawei, they're also banned from selling to it. Huawei "buys from" over 30 U.S. companies, (for example, Windows licenses from Microsoft). CNN reports that U.S. firms could lose up to $11 billion in revenues.

Semiconductor Chip Sales Suffer Fourth Largest Decline in 35 Years

According to the World Semiconductor Trade Statistics (WSTS) organization, the semiconductor manufacturing world has just seen one of the largest contractions in the last 35 years. The downturn on produced revenue for manufacturers for the month of March consolidated into a decline of 1.8% compared to February of this year, and a decline of 13% when compared to March 2018 - but quarter-reviewed revenues were even worse. In greenback terms, the semiconductor industry saw a decline from $114.7 billion in the previous quarter to "just" $96.8 billion.

The decline was across all semiconductor product categories, as John Neuffer, president and CEO of the Semiconductor Industry Association (SIA) trade group, said: "Sales in March decreased on a year-to-year basis across all major regional markets and semiconductor product categories, consistent with the cyclical trend the global market has experienced recently." Market analysis firm IC Insights says that the decline was more severe than the WSTS reports, and that it totaled a 17.1% reduction in revenue for the first quarter of this year, making it the fourth biggest decline since 1984. As IC Insights said in a statement, "The first quarter is usually the weakest quarter of the year for the IC market, averaging a sequential decline of 2.1% over the past 36 years, but the severity of the 1Q19/4Q18 IC market drop has started this year off at a very low level."

TSMC Completes 5 nm Design Infrastructure, Paving the Way for Silicon Advancement

TSMC announced they've completed the infrastructure design for the 5 nm process, which is the next step in silicon evolution when it comes to density and performance. TSMC's 5 nm process will leverage the company's second implementation of EUV (Extreme Ultra Violet) technology (after it's integrated in their 7 nm process first), allowing for improved yields and performance benefits.

According to TSMC, the 5 nm process will enable up to 1.8x the logic density of their 7 nm process, a 15% clock speed gain due to process improvements alone on an example Arm Cortex-A72 core, as well as SRAM and analog circuit area reduction, which means higher number of chips per wafer. The process is being geared for mobile, internet, and high performance computing applications. TSMC also provides online tools for silicon design flow scenarios that are optimized for their 5 nm process. Risk production is already ongoing.

AMD Says Not to Count on Exotic Materials for CPUs in the Next Ten Years, Silicon Is Still Computing's Best Friend

AMD's senior VP of AMD's datacentre group Forrest Norrod, at the Rice Oil and Gas HPC conference, said that while graphene does have incredible promise for the world of computing, it likely will take some ten years before such exotic material are actually taken advantage off. As Norrod puts it, silicon still has a pretty straightforward - if increasingly complex - path down to 3 nanometer densities. And according to him, at the rate manufacturers are being able to scale down their production nodes further, the average time between node transitions stands at some four or five years - which makes the jump to 5 nm and then 3 nm look exactly some 10 years from now, where Norrod expects to go through two additional shrinking nodes for the manufacturing process.

Of course, graphene is being hailed as the next best candidate for taking over silicon's place at the heart of our more complex, high-performance electronics, due, in part, to its high conductivity independent of temperature variation and its incredible switching resistance - it has been found to be able to operate at Terahertz switching speeds. It's a 2D material, which means that implementations of it will have to occur in deposited sheets of graphene across some other material.

Capital Expenditure on Silicon Chip Manufacturing to Rise to $67.5 billion in 2019

The race for smaller fabrication processes has become more and more expensive, and the expenses in R&D and factory retooling only look to increase. This - alongside the expected increase in demand from silicon-embedded products, which are almost all of them - means that additional funding will be poured into chip manufacturing capabilities. A report from SEMI indicates that the 14% increased investment in 2018 to $62.8 billion will increase a further 7.5% next year, reaching capital expenditure of $67.5 billion in 2019.

3D NAND fabrication plants lead the charge in investment, even if the market is facing some issues stemming from oversupply. The demand growth is being taken into account for these new expansion plans, however, with denser and denser chips being required for all manner of products. This is part of the reason why 43% of this years' spending has been allotted to new NAND factories, but the ratio for 2019 is a much lower 19% increase.

Silicon Lottery Posts its Pricing of the Core i9-9900K and i7-9700K

Silicon Lottery is an online retailer that sells computer hardware its employees personally bin to pick out the best performing parts, at higher-than-MSRP prices. It listed its pricing for the upcoming Intel Core i9-9900K 8-core/16-thread processor, and the Core i7-9700K 8-core/8-thread part. The site currently reports both parts as "sold out" either because they've actually sold out all their pre-order inventory, or because they have't built inventories yet. Regardless, the i9-9900K is listed at USD $479.99, and the i7-9700K at $369.99.

We've been actively tracking down possible list prices of Intel's 9th generation Core processors. Our most recent article on the topic predicts the i9-9900K to be priced around $450, the i7-9700K at $350, and the i5-9600K at $250. Either Silicon Lottery's listings don't include any premiums, or Intel could surprise us with prices lower than our predictions.

On The Coming Chiplet Revolution and AMD's MCM Promise

With Moore's Law being pronounced as within its death throes, historic monolithic die designs are becoming increasingly expensive to manufacture. It's no secret that both AMD and NVIDIA have been exploring an MCM (Multi-Chip-Module) approach towards diverting from monolithic die designs over to a much more manageable, "chiplet" design. Essentially, AMD has achieved this in different ways with its Zen line of CPUs (two CPU modules of four cores each linked via the company's Infinity Fabric interconnect), and their own R9 and Vega graphics cards, which take another approach in packaging memory and the graphics processing die in the same silicon base - an interposer.

Q4 2017 300 mm Silicon Wafer Pricing to Increase 20% YoY in DRAM-like Squeeze

Silicon wafers are definitely the best kind of wafers for us tech enthusiasts, but as we all know, required financial resources for the development and production of these is among the most intensive in development costs and R&D. It's not just about the cost of employing enough (and crucially, good enough) engineers that can employ the right tools and knowledge to design the processing miracles that are etched onto wafers; there's also the cost of good, old production as well. Extreme Ultraviolet Lithography Systems that are used for the production of silicon wafers are about the size of a city bus, and typically cost more than 100 million euros ($115.3 million) each. ASML, a Dutch company that specializes in this kind of equipment, announced this year it was expecting to see a 25% revenue growth for 2017. Increased demand for these systems - and added cost of development of ever increasingly small and complex etchings in wafers - means this sector is seeing strong growth. But where there is strong growth, there is usually high demand, and high demand means higher strain on supply, which may sometimes not be able to keep up with the market's needs.

This is seemingly the case for wafer pricing; as demand for wafer production has been increasing, so to are prices. Faced with increased demand, companies are usually faced with a tough question to answer in regards to the correct course of action. Usually, it goes like this: higher demand at the same supply level means higher pricing. However, if supply isn't enough to satisfy demand, manufacturers are losing out on potential increased sales. This leads most companies to increase supply relative to demand, but always with lower projected output than demand requires, so they can bask in both increased ASP (Average Sale Price) and higher number of sales. This has been the case with DRAM memory production for some time now: and is happening with 300 mm silicon wafers as well.

MIT, Stanford Partner Towards Making CPU-Memory BUSes Obsolete

Graphene has been hailed for some time now as the next natural successor to silicon, today's most used medium for semiconductor technology. However, even before such more exotic solutions to current semiconductor technology are employed (and we are still way off that future, at least when it comes to mass production), engineers and researchers seem to be increasing their focus in one specific part of computing: internal communication between components.

Typically, communication between a computer's Central Processing Unit (CPU) and a system's memory (usually DRAM) have occurred through a bus, which is essentially a communication highway between data stored in the DRAM, and the data that the CPU needs to process/has just finished processing. The fastest CPU and RAM is still only as fast as the bus, and recent workloads have been increasing the amount of data to be processed (and thus transferred) by orders of magnitude. As such, engineers have been trying to figure out ways of increasing communication speed between the CPU and the memory subsystem, as it is looking increasingly likely that the next bottlenecks in HPC will come not through lack of CPU speed or memory throughput, but from a bottleneck in communication between those two.

Samsung Announces Comprehensive Process Roadmap Down to 4 nm

Samsung stands as a technology giant in the industry, with tendrils stretching out towards almost every conceivable area of consumer, prosumer, and professional markets. It is also one of the companies which can actually bring up the fight to Intel when it comes to semiconductor manufacturing, with some analysts predicting the South Korean will dethrone Intel as the top chipmaker in Q2 of this year. Samsung scales from hyper-scale data centers to the internet-of-things, and is set to lead the industry with 8nm, 7nm, 6nm, 5nm, 4nm and 18nm FD-SOI in its newest process technology roadmap. The new Samsung roadmap shows how committed the company is (and the industry with it) towards enabling the highest performance possible from the depleting potential of the silicon medium. The 4 nm "post FinFET" structure process is set to be in risk production by 2020.

This announcement also marks Samsung's reiteration on the usage of EUV (Extreme Ultra Violet) tech towards wafer manufacturing, a technology that has long been hailed as the savior of denser processes, but has been ultimately pushed out of market adoption due to its complexity. Kelvin Low, senior director of foundry marketing at Samsung, said that the "magic number" for productivity (as in, with a sustainable investment/return ratio) with EUV is 1,500 wafers per day. Samsung has already exceeded 1,000 wafers per day and has a high degree of confidence that 1,500 wafers per day is achievable.

GlobalFoundries Announces its 12 nm FD-SOI Silicon Fabrication Node

GLOBALFOUNDRIES today unveiled a new 12nm FD-SOI semiconductor technology, extending its leadership position by offering the industry's first multi-node FD-SOI roadmap. Building on the success of its 22FDXTM offering, the company's next-generation 12FDXTM platform is designed to enable the intelligent systems of tomorrow across a range of applications, from mobile computing and 5G connectivity to artificial intelligence and autonomous vehicles.

As the world becomes more and more integrated through billions of connected devices, many emerging applications demand a new approach to semiconductor innovation. The chips that make these applications possible are evolving into mini-systems, with increased integration of intelligent components including wireless connectivity, non-volatile memory, and power management-all while driving ultra-low power consumption. GLOBALFOUNDRIES' new 12FDX technology is specifically architected to deliver these unprecedented levels of system integration, design flexibility, and power scaling.
Return to Keyword Browsing