Prediction articles:
2020: A consumer electronics forecast for the year(s) ahead
AI Chips: What Will 2020 Bring?
CEO Outlook: 2020 Vision: 5G, China and AI are prominent, but big changes are coming everywhere
Top 10 Tech Failures From 2019 That Hint At 2020 Trends – Last year’s tech failures often turn into next year’s leading trends
Trends:
AMD’s 7nm Ryzen 4000 CPUs are here to take on Intel’s 10nm Ice Lake laptop chips
Top 9 challenges IT leaders will face in 2020: From skills shortages to privacy concerns
From the oil rig to the lake: a shift in perspective on data
In December 2020, the new IEC/EN 62368-1 will replace the existing safety standards EN 60950-1 and EN 60065-1
Use of technology money outside company IT department is the new normal
Tech to try:
12 Alternative Operating Systems You Can Use In 2020
CONTINUOUS INTEGRATION: WHAT IT IS AND WHY YOU NEED IT
Research:
Universal memory coming? New type of non-volatile general purpose memory on research, some call it UltraRAM.
1,318 Comments
Tomi Engdahl says:
Start Monitoring AWS in Minutes
Create customizazble dashboards for AWS, EC2, ECS, Lambda, and 400+ other technologies with Datadog.
https://www.datadoghq.com/dg/monitor/aws-dashboards-vd/
Tomi Engdahl says:
Swift: Google’s bet on differentiable programming
https://tryolabs.com/blog/2020/04/02/swift-googles-bet-on-differentiable-programming/
started working on making Swift the first mainstream language with first-class language-integrated differentiable programming capabilities. The scope and initial results of the project have been remarkable, and general public usability is not very far off.
Despite this, the project hasn’t received a lot of interest in the machine learning community and remains unknown to most practitioners. This can be attributed in part to the choice of language, which has largely been met with confusion and indifference, as Swift has almost no presence in the data science ecosystem and has mainly been used for building iOS apps.
What is wrong with you, Python?!
Python is by far the most used language in machine learning, and Google has a ton of machine learning libraries and tools written in it. So, why Swift? What’s wrong with Python?
To put it bluntly, Python is slow. Also, Python is not great for parallelism.
To get around these facts, most machine learning projects run their compute-intensive algorithms via libraries written in C/C++/Fortran/CUDA, and use Python to glue the different low-level operations together. For the most part, this has worked really well, but as with all abstractions, it can create some problems. Let’s go over some of those.
External binaries
Calling external binaries for every compute-intensive operation limits developers to working on a small portion of the algorithm’s surface area. Writing a custom way to perform convolutions, for example, becomes off limits unless the developer is willing to step down into a language like C. Most programmers choose not to do so, either because they have no experience with writing low level performant code, or because jumping back and forth between Python’s development environment and some low level language’s environment becomes too cumbersome to justify.
Tomi Engdahl says:
Every developer has a test environment, some are lucky enough to have a separate production environment…
https://www.facebook.com/nixcraft/videos/debugging-code-in-production/702846050452216/
Tomi Engdahl says:
Reliability Challenges Grow For 5/3nm
https://semiengineering.com/reliability-challenges-grow-for-5-3nm/
New transistors, materials and higher density are changing the testing paradigm.
Ensuring that chips will be reliable at 5nm and 3nm is becoming more difficult due to the introduction of new materials, new transistor structures, and the projected use of these chips in safety- and mission-critical applications.
Each of these elements adds its own set of challenges, but they are being compounded by the fact that many of these chips will end up in advanced packages or modules, in applications where stress based on heat, vibration and other physical effects are no longer completely predictable.
Tomi Engdahl says:
// BUGS: None known, many assumed.
Tomi Engdahl says:
Michael McWhertor / Polygon:
Google makes Stadia free to use for anyone with a Gmail address and will give new users two months of Stadia Pro for free — Stadia Pro is also free for two months — Google’s video game streaming platform, Stadia, is now free to anyone with a Gmail address, the company announced on Wednesday.
Google Stadia now free to anyone with a Gmail addres
Stadia Pro is also free for two months
https://www.polygon.com/2020/4/8/21213551/google-stadia-free-pro-subscription
Google’s video game streaming platform, Stadia, is now free to anyone with a Gmail address, the company announced on Wednesday. To sweeten the deal, Google is also giving new users two months of Stadia Pro — including access to nine games — for free.
Existing Stadia Pro subscribers won’t be charged for the next two months of the service, Google said.
Previously, access to Stadia required purchasing the $129 Google Stadia Premiere Edition, a bundle that includes a Chromecast Ultra, a wireless Stadia Controller, and three months of Stadia Pro, the service that offered free games and video streams up to 4K resolution and 60 frames per second with HDR lighting.
As of April 8, access to the base-level version of Stadia — games stream at a maximum resolution of 1080p — will be free by signing up at the Stadia website. Users still have to purchase games to own them, but those games can be played on a PC, Chrome OS tablet, Google Pixel phone, and other supported Android devices. A Stadia Controller (which can be purchased separately for $69) is not required; users can also play with a supported USB controller or mouse and keyboard.
After the two-month trial period, Stadia Pro will cost $9.99 per month. Subscribers can cancel their subscriptions online, if they choose to do so.
Tomi Engdahl says:
Mary Jo Foley / ZDNet:
Sources: Microsoft won’t be delivering its Windows 10X-powered Surface Neo dual-screen devices this year, nor will it ship Windows 10X to third-party OEMs
Microsoft: Don’t expect any Windows 10X devices this calendar year
https://www.zdnet.com/article/microsoft-dont-expect-any-windows-10x-devices-this-calendar-year/
Microsoft is shifting its Windows focus to reflect the new world where virtualization and single-screen devices are more of a priority than brand-new form factors.
Tomi Engdahl says:
COVID-19: Economic and Microelectronics Industry Impacts – Insights from McKinsey & Company
https://blog.semi.org/technology-trends/covid-19-economic-and-microelectronics-industry-impacts-insights-from-mckinsey-company
For five days in the latter half of March, the pall of the heavy human and economic toll COVID-19 has exacted in China appeared to be lifting. The epicenter of Wuhan reported no new coronavirus infections through domestic transmission. And in an initial step to loosen its nationwide lockdown, China began reversing restrictions on travel within its borders.
Now, in another sign of progress, the region’s idled factory workforce is preparing to return to the production lines. Outside of Hubei province, home to Wuhan, most manufacturing workers are expected to be back on the job by the end of this month
McKinsey is also “seeing evidence of a rebound in demand led by China’s online sales” as rising consumer confidence and a surge in the popularity of work-from-home policies spur strong spending on laptop computers, Chenneveau said.
The turnaround stands in stark contrast to the unprecedented drop in demand McKinsey saw across retail and durable goods in China early in the year. Over the first two months, passenger car sales plunged 90 percent, smart phone receipts 40 percent and retail sales 21 percent, leading to what Chenneveau calls a whiplash effect that could disrupt supply chains as manufacturers and shipping companies scramble to meet pent-up demand once a recovery takes hold.
And while China is buoyed by the prospect of normalizing its workforce and manufacturing capabilities, parts shortages are bottlenecking production. In the United States and Europe, where 60 percent of air freight is carried in cargo holds of passenger aircraft, logistics concerns loom large with the widespread flight groundings.
“Logistics must be a priority in any crisis war room because it’s a big challenge,”
In Asia, the semiconductor supply chain is working to overcome intractable challenges caused by COVID-19 including sourcing raw materials for chip manufacturing and maintaining assembly and test operations
a shortage of fab operators and engineers. Downstream, the inability to package, test and qualify products risks exacerbating the supply constraints.
Patel said another acute challenge is that most semiconductor manufacturers and suppliers are operating under restricted practices, making it harder to sustain engineering activities vital to new product introductions, new process development and capital equipment expansion. In the longer term, the supply chain fallout hold implications for product life cycles and investments in capacity and next-generation technology – factors that analysts will need to monitor in evaluating the economic impact.
Returning Workers Key to Economic Recovery
Issuing shelter-in-place orders have been an effective antidote to the spread of COVID-19 but a double-edged sword as nations worldwide sustain the economic blowback. Discretionary consumer spending on items such as automobiles has dropped by 45 percent globally so far this year, business investment has fallen and trade has seen a sharp slowdown, said Sven Smit, Chairman and Director at the McKinsey Global Institute, speaking at the webinar.
A lockdown for as little as a month can slash aggregate global GDP by as much as 10 percent, a scenario McKinsey expects to play out in the second quarter of 2020. The drop would be the deepest since World War II and larger than the plunge in the first quarter of the Great Depression, raising the question of how long governments can afford to keep workers holed up at home.
“The economic shock is unprecedented,” Smit said. “We’ve never sent people home to not work. Even in World War II, next to the front lines, people were harvesting food.”
China offers a potential blueprint for economic recovery. McKinsey estimates that China’s rigorous containment efforts could help its economy bounce back in as little as six months – a V-shaped rebound. Western nations generally have not been as forceful with their containment measures. For them, the fight against the pathogen could be prolonged, deepening the economic damage.
Yet even with the best protective lockdowns, a new challenge arises: The longer shelter-in-place orders remain in effect to contain the spread of the virus, the longer the economic impact drags on. “Until the path to return to work becomes clearer, people will not be confident to spend,” Smit said.
Confronted with that reality, governments worldwide must strike the delicate balance between safeguarding the lives of people – critical forces of economic growth through consumer spending – and limiting the economic shock. The faster the virus can be brought to heel, the softer the impact to economies around the world. And the stronger the return-to-work protocols in place once COVID-19 has been brought under control, the faster workers can get back to their jobs. Smit believes resolving both issues simultaneously is not only possible but necessary for a return to normalcy.
Tomi Engdahl says:
https://nora.codes/post/stop-making-students-use-eclipse/
Tomi Engdahl says:
How Azure can provide a secure remote access without VPN
Apr 8, 2020, 11:45 AM
https://www.nitor.com/en/news-and-blogs/how-azure-can-provide-secure-remote-access-without-vpn
Enabling your users to work efficiently from home is pretty much a must-have requirement for all modern IT-infrastructures, and after the coronavirus hit us, more and more people need remote working possibilities. MS Azure offers us some alternatives to traditional VPN connections
Tomi Engdahl says:
5 open source activities while you work from home
https://opensource.com/article/20/4/open-source-activities
Work-life balance is important whether you’re working from home temporarily or it’s your usual mode of earning a living.
Tomi Engdahl says:
https://www.blackhillsinfosec.com/how-to-create-a-soho-router-using-ubuntu-linux/
Tomi Engdahl says:
https://talented.fi/fi/markkinatilannekuva/
Tomi Engdahl says:
Xbox Series X Graphics Source Code Reportedly Stolen, Being Held for $100 Million Ransom
The leaked data refers to the rumoured Arden GPU.
https://nordic.ign.com/hardware/35122/news/xbox-series-x-graphics-source-code-reportedly-being-held-to-100-million-ransom-by-hacker
Tomi Engdahl says:
Topological Photonics: What It Is and Why We Need It
https://spectrum.ieee.org/semiconductors/optoelectronics/topological-photonics-what-it-is-why-we-need-it
Since topological insulators were first created in 2007, these novel materials, which are insulating on the inside and conductive on the outside, have intrigued researchers for their potential in electronics. However, a related but more obscure class of materials—topological photonics—may reach practical applications first.
Tomi Engdahl says:
Squeezing Light Out of Silicon
https://spectrum.ieee.org/tech-talk/semiconductors/optoelectronics/squeezing-light-out-of-silicon
Tomi Engdahl says:
https://www.aineetonkulttuuriperinto.fi/fi/artikkeli/elavan-perinnon-kansalliseen-luetteloon-12-uutta-kohdetta-2
Tomi Engdahl says:
Demoskene, sahansoitto ja romanien hevostaidot – Elävän perinnön kansalliseen luetteloon 12 uutta kohdetta
https://www.museovirasto.fi/fi/ajankohtaista/demoskene-sahansoitto-ja-romanien-hevostaidot-elavan-perinnon-kansalliseen-luetteloon-12-uutta-kohdetta
Tomi Engdahl says:
AutoHotkey
Powerful. Easy to learn.
https://www.autohotkey.com/
The ultimate automation scripting language for Windows.
Tomi Engdahl says:
https://www.cs50.harvard.edu/x/2020/
Tomi Engdahl says:
Gartner Forecasts Worldwide Semiconductor Revenue to Decline 0.9% in 2020 Due to Coronavirus Impact
Expected Memory Revenue Growth of 13.9% Could Help Prevent Steep Decline
https://www.gartner.com/en/newsroom/press-releases/2020-04-09-gartner-forecasts-worldwide-semiconductor-revenue-to-0
Due to the impact of the coronavirus on semiconductor supply and demand, worldwide semiconductor revenue is forecast to decline 0.9% in 2020, according to Gartner, Inc. This is down from the previous quarter’s forecast of 12.5% growth.
“The wide spread of COVID-19 across the world and the resulting strong actions by governments to contain the spread will have a far more severe impact on demand than initially predicted,” said Richard Gordon, research practice vice president at Gartner. “This year’s forecast could have been worse, but growth in memory could prevent a steep decline.”
Tomi Engdahl says:
https://semiengineering.com/week-in-review-manufacturing-test-92/
For example, IC Insights has lowered its IC forecast from 3% to minus 4% in 2020. The IC market is now expected to hit $345.8 billion in 2020, which is $39.0 billion less than the original forecast, which called for an 8% increase this year.
Due to the impact of the coronavirus on semiconductor supply and demand, worldwide semiconductor revenue is forecast to decline 0.9% in 2020, according to a new report from Gartner. This is down from the previous quarter’s forecast of 12.5% growth. “The wide spread of COVID-19 across the world and the resulting strong actions by governments to contain the spread will have a far more severe impact on demand than initially predicted,” said Richard Gordon, research practice vice president at Gartner. “This year’s forecast could have been worse, but growth in memory could prevent a steep decline.”
https://www.icinsights.com/news/bulletins/Global-IC-Market-Forecast-Lowered-From-3-To-4/
Tomi Engdahl says:
https://www.businesswire.com/news/home/20200409005730/en/Strategy-Analytics-COVID-19-Drives-Recession-Damaging-Automotive
Covid-19 will also disrupt the automotive, consumer electronics and IT infrastructure businesses worldwide in 2020, according to Strategy Analytics. Real GDP is predicted to fall, according to the firm, who also sees a recovery in 2021. “The entire supply chain for digital products in our homes, our cars, and associated with our mobile lifestyles will experience significant damage during this downturn that is likely to be felt globally over the next three to four quarters,” said Harvey Cohen, president of Strategy Analytics.
Strategy Analytics: COVID-19 Drives Recession Damaging Automotive, Consumer Electronics and Semiconductors Globally
https://www.businesswire.com/news/home/20200409005730/en/Strategy-Analytics-COVID-19-Drives-Recession-Damaging-Automotive
Real GDP Predicted to fall 10% to 15% in 2020 as consumer spending plummets.
Tomi Engdahl says:
Experts Invent a New Way to Track Semiconductor Technology
https://spectrum.ieee.org/nanoclast/semiconductors/devices/experts-invent-a-new-way-to-track-semiconductor-technology
A group of some of the most noted device engineers on the planet, including several IEEE Fellows and this year’s IEEE Medal of Honor recipient, is proposing a new way of judging the progress of semiconductor technology. Today’s measure, the technology node, began to go off the rails almost two decades ago. Since then, the gap between what a technology node is called and the size of the devices it can make has only grown. After all, there is nothing in a 7-nanometer chip that is actually that small. This mismatch is much more of problem than you might think, argues one of the group, Stanford University professor H.-S. Philip Wong.
“The broader research community has a feeling that [device] technology is over,” he says. “Nothing could be further from the truth.”
. “We’re at this juncture where the path to the future doesn’t seem to be very clear. 2D scaling is coming to an end, but we all know that there is a future,” says Wong. “So how do we quantify that?”
Their answer is a set of three numbers, which unlike today’s system, increase as technology progresses rather than decreasing. Taken together, the three metrics describe a technology’s impact on a computing system as a whole.
Tomi Engdahl says:
Big Changes In Tiny Interconnects
https://semiengineering.com/big-changes-in-tiny-interconnects/
Below 7nm, get ready for new materials, new structures, and very different properties.
One of the fundamental components of a semiconductor, the interconnect, is undergoing radical changes as chips scale below 7nm.
Some of the most pronounced shifts are occurring at the lowest metal layers. As more and smaller transistors are packed onto a die, and as more data is processed and moved both on and off a chip or across a package, the materials used to make those interconnects, the structures themselves, and the entire approach for utilizing those structures is changing.
The problem is that copper, which has been used been used for those interconnects since 130nm, has largely run out of steam. So at 10nm, Intel made a switch. The local interconnect layers—M0 and M1—incorporate cobalt, not copper
Generally, copper is still used for the other and traditional backend-of-the-line (BEOL) layers. Cobalt is mainly used for the liners here, although ruthenium is gaining steam. Other materials are also being explored as well as a newer technology called buried power rails.
“There are some really interesting interconnect developments, including the use of different materials and deposition techniques for those materials,” said David Fried, vice president of computational products at Lam Research/Coventor. “This includes thinner liners, barriers and seed layers, along with more conformal deposition techniques. We are also seeing different metals with lower resistance being used, and different combinations of liners, barriers, seed and fill that end up with a lower aggregate resistance. There are also many developments ongoing with dielectrics and lowering the ‘K’ but keeping them mechanically stable. There is a tremendous amount of work on plating versus depositing versus CVD, and new materials and new process technologies are being applied to new materials.”
This is much more complicated than it sounds. One of the problems with copper is that at the most advanced nodes it can diffuse into surrounding materials. That requires a barrier layer, but as scaling continues to 5nm and 3nm, those barrier layers need to be thinner. But they also need to be conductive.
“You’re trying to put a barrier layer down at 1nm or 2nm, so for that film to be reliable it has to be very consistent,”
Tomi Engdahl says:
Making Chips At 3nm And Beyond
https://semiengineering.com/making-chips-at-3nm-and-beyond/
Lots of new technologies, problems and uncertainty as device scaling continues.
Select foundries are beginning to ramp up their new 5nm processes with 3nm in R&D. The big question is what comes after that.
Work is well underway for the 2nm node and beyond, but there are numerous challenges as well as some uncertainty on the horizon.
Tomi Engdahl says:
Gradual Rebound or Slight Dip – Two Scenarios for COVID-19 Impact to 2020 Global Silicon Wafer Market Sales
https://blog.semi.org/business-markets/gradual-rebound-or-slight-dip-two-scenarios-for-covid-19-impact-to-2020-global-silicon-wafer-market-sales
Uncertainty has gripped the silicon wafer market as the COVID-19 pandemic threatens to upend growth projections for 2020. Declines in both shipments and revenue plagued the silicon wafer market in 2019, a downturn that had given way to optimism for 2020 with rising expectations for normalizing inventory levels, memory market improvements, data center market growth and the 5G market takeoff.
Tomi Engdahl says:
Top 10 Critical Subsystems Suppliers of 2019: EUV Shines Light on Optical Subsystems Suppliers
https://blog.semi.org/business-markets/top-10-critical-subsystems-suppliers-of-2019-euv-shines-light-on-optical-subsystems-suppliers
Tomi Engdahl says:
https://blog.semi.org/business-markets/coronavirus-takes-on-electronics-industry
Tomi Engdahl says:
EDN in 1981:
“A new production system will emerge … and with it a new emphasis on the home as the center of society—a home with a low-cost workstation, with a ‘smart’ typewriter, perhaps, along with a facsimile machine or computer console and teleconferencing equipment.” #electroniccottage #FutureShock #WFH #StayHome https://buff.ly/3agWaL7
Tomi Engdahl says:
TSMC Q1 net profit hits new high
https://focustaiwan.tw/business/202004160011
Taipei, April 16 (CNA) Taiwan Semiconductor Manufacturing Co. (TSMC), the world’s largest contract chipmaker, on Thursday reported that its net profit for the first quarter of this year hit a new quarterly high in the company’s history, with market analysts attributing the strong showing to solid demand for emerging technologies such as 5G and high-performance computing (HPC) devices.
5G phones key to chip market recovery
https://www.digitimes.com/news/a20200416PD203.html
The coronavirus pandemic has had an adverse impact on the global IC market, particularly the smartphone segment, but sales of 5G smartphones will be key to the overall demand recovery in 2020, according to sources at Taiwan-based IC distributors.
The market for 5G smartphones is expected to expand in 2020, with the availability of 5G-capable iPhone set to spur the market demand.
Nevertheless, component suppliers engaged in the supply chain for smartphones are increasingly uncertain about whether the deliveries for next-generation 5G models can be on schedule, the sources said. For Apple’s 5G iPhone, related chip and component production should kick off by June 2020 followed by device assembly between July and August in order for it to be unveiled in September, the sources said.
Uncertainty has emerged about whether Apple will debut its flagship iPhone series in the usual September timeframe this year, judging from the severity of the coronavirus pandemic and weak end-market demand, the sources noted.
Tomi Engdahl says:
Scaling At The Angstrom Level
Shrinking features will continue, but not everywhere and not all the time.
https://semiengineering.com/scaling-at-the-angstrom-level/
It now appears likely that 2nm will happen, and possibly the next node or two beyond that. What isn’t clear is what those chips will be used for, by whom, and what they ultimately will look like.
The uncertainty isn’t about the technical challenges. The semiconductor industry understands the implications of every step of the manufacturing process down to the sub-nanometer level, including how to create new materials that can withstand a narrow range of temperatures or disappear entirely without a trace.
The real problem is cost, and how economies of scale will play out in the future. From the standpoint of being able to design and manufacture chips that work, the industry appears to have some pretty good options for pushing on to the next three or four nodes. From the standpoint of commercial viability, there are some looming unanswered questions.
Tomi Engdahl says:
https://semiengineering.com/week-in-review-manufacturing-test-93/
TSMC posted mixed results for the quarter, although there was a capital spending surprise. “It maintained its 2020 capex at $15B-$16B despite smartphone softness, primarily to support a strong 5nm ramp, led by demand from 5G and HPC customers,” said Weston Twigg, an analyst at KeyBanc, in a research note. “Despite lowering its industry outlook, TSMC still expects to grow its own revenue mid- to high-teens this year, well above the industry level.”
TSMC’s 7nm+ process, the first node that used EUV lithography, has entered its second year of production, according to KeyBanc. “6nm, which adds one EUV layer over 7nm+, has entered production and is on track for volume production before the end of 2020,” Twigg said. “5nm technology, which uses EUV lithography more extensively, is already in volume production with good yields, and TSMC continues to expect a smooth and fast ramp in 2H20, driven by demand for mobile and HPC applications.”
As expected, TSMC will extend the finFET to 3nm. “TSMC noted its 3nm development is on track, with risk production scheduled in 2021 and volume production targeted for 2H22. The 3nm node is expected to be a full node transition from 5nm. TSMC noted that 3nm will remain a finFET-based design,” Twigg added.
Tomi Engdahl says:
It’s been a difficult period for the semiconductor industry. The coronavirus outbreak has put a damper on what was supposed to be a strong year in the semiconductor industry in 2020. Many are holding out hopes for a rebound in the second half of the year. That’s still a big unknown. The forecasts are gloomy.
Taking A Pulse On The IC Biz
https://semiengineering.com/taking-a-pulse-on-the-ic-biz/
Experts discuss the status of the IC/equipment industries.
Global Semiconductor Capex Forecast to Drop 3% Again This Year
Using its “baseline” assumptions shown in the soon-to-be-released April Update to the 2020 edition of The McClean ReportA Complete Analysis and Forecast
Global Semiconductor Capex Forecast to Drop 3% Again This Year
Semiconductor producers hoping to keep capital spending plans intact despite virus outbreak.
https://www.icinsights.com/news/bulletins/Global-Semiconductor-Capex-Forecast-To-Drop-3-Again-This-Year/
Tomi Engdahl says:
TSMC 3nm process development remains on track
TSMC has disclosed that its 3nm technology development remains on track, with risk production scheduled for 2021 followed by volume production
https://www.digitimes.com/news/a20200417PD202.html
Tomi Engdahl says:
Scaling CMOS Image Sensors
Manufacturing issues grow as cameras become more sophisticated.
https://semiengineering.com/scaling-cmos-image-sensors/
After a period of record growth, the CMOS image sensor market is beginning to face some new and unforeseen challenges.
CMOS image sensors provide the camera functions in smartphones and other products, but now they are facing scaling and related manufacturing issues in the fab. And like all chip products, image sensors are seeing slower growth amid the coronavirus outbreak.
Manufactured at mature nodes in 200mm and 300mm fabs, these sensors are used in phones, cars, consumer products, industrial/medical systems, and security cameras. Smartphones, for example, incorporate two or more cameras, each of which is powered by a CMOS image sensor that converts light into signals in to order to create images.
For example, Samsung’s new 5G smartphone consists of five cameras, including a rear-facing, wide-angle camera based on a 108-megapixel (MP) image sensor. This equates to over 100 million pixels on a small die size. A front-facing camera for selfies incorporates a 48MP image sensor based on the world’s smallest pixel pitch—0.7µm, according to TechInsights.
An image sensor incorporates a multitude of tiny photosensitive pixels.
“As higher-bandwidth data performance progressed from 3G to 4G and now to 5G, the demand for higher quality cameras has grown,”
Still, image sensor vendors face some challenges. For years they have been racing to reduce the pixel pitch. That way they can pack more pixels in an image sensor, which boosts the resolution of the device. Recently, though, pixel scaling is becoming more difficult, as the pitch approaches the wavelength of light. “Pixel R&D teams now have to find new ways to avoid reduction in sensitivity and more crosstalk in the sensor,” said Lindsay Grant, vice president of process engineering at OmniVision.
Nonetheless, image sensor vendors have found ways to solve some of the challenges. Among them:
New processes. High-k films and other fab technologies have jumpstarted pixel scaling.
Die stacking and interconnects. Putting different functions on two dies and stacking them isn’t new. But new interconnect schemes, such as pixel-to-pixel connections, are in R&D.
Image sensor market dynamics
There are two main types of image sensors — CMOS image sensors and charge-coupled devices (CCDs). CCDs, which are current-driven devices, are found in digital cameras and various high-end products.
CMOS image sensors are different. “A complementary metal oxide semiconductor (CMOS) image sensor has a photodiode and a CMOS transistor switch for each pixel, allowing the pixel signals to be amplified individually,” according to TEL’s website.
Targeted for various applications, CMOS image sensors come in different formats, frame rates, pixel sizes and resolutions. Image sensors have global or rolling shutters.
Suppliers are split into two camps—fabless and IDMs. IDMs have their own fabs, while fabless companies use foundries. In either case, a vendor manufactures image sensor dies on a wafer, which are cut and assembled into a package.
Some 65% of image sensors are produced in 300mm fabs, according to Yole Développement. “200mm is still critical for a wide range of security, medical and automotive CMOS image sensor products,”
Today, Sony is the largest supplier of CMOS image sensors, followed by Samsung and OmniVision. Other suppliers include Sharp, ON Semi, STMicroelectronics, GalaxyCore, SK Hynix, Panasonic and Canon, according to IC Insights.
In 2019, image sensor sales reached $18.4 billion, up 30% over 2018, according to IC Insights. “In 2020, we currently forecast a 3% drop in CMOS image sensor sales to $17.8 billion, snapping the string of record sales because of a falloff in demand for sensors in cellphones and other systems in the Covid-19 virus health crisis,” said Rob Lineback, an analyst with IC Insights.
The big driver is smartphones. In 2018, there were 2.5 cameras per phone, according to Yole. “In 2019, it has jumped to 2.8 cameras per smartphone. We see that it will go to three cameras per smartphone in 2020,” said Guillaume Girardin, an analyst at Yole.
Higher-resolution cameras don’t necessarily equate to better photos. “It’s a tradeoff question between the pixel size and resolution,” Girardin said. “Pixel scaling means that it has more pixels. When the resolution is more than 40MP and 50MP, the capabilities may be beyond the human eye to see what they capture. For CMOS image sensors, the pixel with a better quantum efficiency (QE) and a signal-to-noise ratio are the most important things for image quality.”
In addition, smartphones will not displace DLSR cameras for the professional. But clearly, smartphones offer more features than ever before.
In other innovations, vendors are shipping near-infrared (NIR) image sensors. NIR, which illuminates objects with wavelengths outside the visible spectrum, is designed for applications that operate in near or total darkness. OmniVision’s new NIR technology provides a 25% improvement in the invisible 940nm NIR light spectrum and a 17% bump at the barely visible 850nm NIR wavelength.
In a separate development, Sony and Prophesee have developed an event-based vision sensor. Targeted for machine vision apps
Several years ago, CMOS image sensor vendors started the so-called pixel scaling race. This refers to the “pixel pitch,”
Higher pixel density equates to more resolution, but not all sensors require smaller pitches.
Years ago, the pixel pitch for an image sensor was at the 7µm generation. Vendors have reduced the pitch along the way, but there have been some hiccups.
An image sensor with a larger pixel size collects more light, which equates to a stronger signal. Larger image sensors take up board space. Image sensors with smaller pixels collect less light, but you can pack more of them on a die. This, in turn, boosts the resolution.
up until 2009, the mainstream CMOS image sensor was based on a frontside-illuminated (FSI) pixel array architecture. In operation, light hits the front side of the device. The microlens gathers the light and transports it to a color filter. Light goes through a stack of interconnects and is captured by a diode. The charge is converted to a voltage at each pixel and then the signals are multiplexed.
Over the years, the FSI architecture enabled vendors to reduce the pitch for several generations. For example, vendors reduced the pitch from 2.2µm in 2006 to 1.75µm in 2007, according to TechInsights.
In 2008, the industry hit the wall with the FSI architecture at the 1.4µm generation. So starting around 2009, vendors moved to a new architecture — backside illumination (BSI). The BSI architecture turns the image sensor upside down. Light enters from the back side of the silicon substrate. The photons have a shorter path to the photodiodes, which boosts the quantum efficiency.
BSI also jump-started pixel scaling. “In terms of pixel scaling, BSI sensor technology allows for optimum pixel dimensions in the range of 1.2µm to 1.4µm, and stacked BSI allows the footprint of sensors with such pixel dimensions to remain below 30mm2,” Lam’s Haynes said. “Pixels with sub-micron dimensions can be enabled using quad-pixel architectures, enabling resolutions in excess of 48MP.”
So at 1.4µm around 2010, the industry moved to another innovation in the fab — deep trench isolation (DTI).
it took vendors three years to move from 1.4µm (2008) to 1.12µm (2011), four years to reach 1µm (2015), and another three to reach 0.9µm (2018), according to TechInsights.
In 2018, Samsung broke the 1µm barrier with 0.9µm, followed by Sony with 0.8µm in 2019, and Samsung with 0.7µm in 2020.
For sub-µm pixel scaling, the industry requires more innovations. “As pixels shrink, thicker active (silicon) is required to maintain a suitable photodiode size,” Fontaine said in a recent presentation. “A key technology enabler for thicker active (silicon) is DTI and associated high-k defect passivation films.”
Will pixel scaling continue? “It’s likely that pixel scaling will continue beyond 0.7µm,” Grant said. “As pixels shrink beyond 0.7µm, many aspects need to be optimized.
Another issue is that the pixel pitch for mobile sensors is approaching the wavelength of light. “Some people may consider this a limit for minimum pixel size,” Grant said. “For example, the 0.6µm pixel pitch is used in R&D today. This is smaller than the wavelength of red light at 0.65µm (650nm). So the question may arise, ‘Why shrink to sub-wavelength? Will there be any useful benefit for the camera user? Shrinking the pixel size to sub-wavelength does not mean there is no valuable spatial resolution information at the pixel level.’”
“The limitation for continued shrink may come from the user benefit rather than the technology. Today, applications continue to find end user value in shrinking the pixel size, so this is driving the trend. As long as that continues, CMOS image sensor technology development will support that direction.”
Besides pixel scaling, CMOS image sensors are undergoing other innovations like die stacking. Vendors are also using different interconnect technologies, such as through-silicon vias (TSVs), hybrid bonding, and pixel-to-pixel.
2012, when Sony introduced a two-die stacked image sensor. Die stacking enables vendors to split the sensor and processing functions on different chips.
For this, Sony developed one pixel-array die, based on a 90nm process. That die was stacked on a separate 65nm image signal processor (ISP) die, which provides the processing functions. The two dies are then connected.
Eventually, others moved to a similar die stack approach. Generally, the top pixel array die is based on mature nodes. The bottom ISP die ranges from 65nm, 40nm and 28nm processes. 14nm finFET technology is in R&D.
Embedded DRAM enables faster data readouts.
Both TSVs and hybrid bonding enable fine pitches. “With respect to stacking of CMOS image sensor pixel and logic wafers, TSV integration and hybrid bonding are likely to continue to co-exist for stacked BSI,” Lam’s Haynes said. “But as multi-stacked BSI sensors become more common place, TSV integration will become increasingly relevant.”
The next big thing is pixel-to-pixel interconnects.
“It allows pixel-level interconnect between each pixel of the sensor and an associated A/D converter. This allows parallel A/D conversion for all pixels.
This architecture supports massive parallel signal transfer, making it possible to read and write all pixel data of the image sensor at high speeds. “It enables global shutter with scaled pixels for real-time, high-resolution imaging for various timing critical applications, such as autonomous vehicles, medical imaging and high-end photography,” Nuruzzaman said.
Tomi Engdahl says:
Making Chips At 3nm And Beyond
Lots of new technologies, problems and uncertainty as device scaling continues.
https://semiengineering.com/making-chips-at-3nm-and-beyond/
Tomi Engdahl says:
Windows 10:n suurpäivitys tulee toukokuussa – sitä pääsee kokeilemaan vaikka heti, jos uskaltaa
https://www.tivi.fi/uutiset/windows-10n-suurpaivitys-tulee-toukokuussa-sita-paasee-kokeilemaan-vaikka-heti-jos-uskaltaa/562dc12e-3136-4366-9d0a-f87ee9c0e216
Tomi Engdahl says:
https://hackaday.com/2020/04/20/cobol-isnt-the-issue-a-misinterpreted-crisis/
Or in the case of COBOL, it’s more that the areas where it makes sense to use the language are limited and highly specific. COBOL’s natural domain is reliable high volume data processing and it’s a perfect language for that use. COBOL is solely outdated as a general-purpose language.
So maybe the issue of finding COBOL programmers isn’t really about the language, but rather what its use implies about the systems using it, and the work environment one can expect. The thing is, those systems are a product of their times, where development processes and methodologies as we know them today simply didn’t exist, embedded in industries where software is merely seen as a tool, if not a nuisance. But COBOL itself is irrelevant here, we’d be in the same situation with any other language. Worse yet, if we don’t start to finally tackle the actual issues at hand, or at least learn from them, and continue to blame everything on an old language, history is going to repeat itself, and we will end up in this exact situation with other languages as well.
Tomi Engdahl says:
Python 2:n tuki päättyy 11 vuoden jälkeen – viimeisessä versiossa ylläpidon mukaan ”erityistä” matemaattista huumoria
https://www.tivi.fi/uutiset/tv/31c8e85b-ffaf-4921-a236-c2837662bb28
Ohjelmointikieli Pythonin 2-sarjan versioiden tuki päättyy, ilmoittivat kielen kehittäjät maanantaina.
Tomi Engdahl says:
The burgeoning field of topological photonics could help move data across chips at terabit-per-second speeds.
Novel Materials Could Help Terahertz Chips Deliver Data at Terabits-Per-Second Rates
https://spectrum.ieee.org/nanoclast/computing/hardware/terahertz-chip
Novel materials known as photonic topological insulators could one day help terahertz waves send data across chips at unprecedented speeds of a trillion bits per second, a new study finds.
Terahertz waves fall between optical waves and microwaves on the electromagnetic spectrum. Ranging in frequency from 0.1 to 10 terahertz, terahertz waves could be key to future 6G wireless networks.
Conventional terahertz waveguides are vulnerable to fabrication defects and considerable signal loss at sharp bends. Now, researchers find the burgeoning field of topological photonics may help solve these problems.
Topology is the branch of mathematics that explores what features of shapes can survive deformation.
Tomi Engdahl says:
Where’s my check? COBOL’s role in delay of stimulus and unemployment payments
With job losses and income plummeting thanks to the coronavirus pandemic we need money as fast as we can, but government systems written in the old COBOL computer language are blocking us from our cash.
https://www.zdnet.com/article/wheres-my-check-cobol-unemployment-and-taxes/
“COBOL is not the issue,” said Derek Britton, product director, mainframe solutions for Micro Focus, the company behind COBOL.
The real problem is that state governments in New Jersey, New York and Connecticut, among others — as well as the IRS — have not kept their codebase up to date. But it’s not entirely their fault either.
The underlying problem isn’t COBOL, or even a lack of COBOL programmers. It’s that no one wants to spend money on infrastructure — be it roads or code — until it’s broken. Well, these programs are now busted.
Tomi Engdahl says:
https://www.zdnet.com/article/google-backed-go-programming-language-developers-reveal-its-top-missing-features/
Tomi Engdahl says:
Kerching! Intel PC chip shortage over just in time for everyone to buy computers for pandemic home working
Sales, profit soar in first three months of 2020 as coronavirus helps Chipzilla print money
https://www.theregister.co.uk/2020/04/23/intel_q1_2020/
Tomi Engdahl says:
Covid-19 Spurs 8 Strategies to Boost China’s Major-Appliance Sales
https://www.eetimes.com/covid-19-spurs-8-strategies-to-boost-chinas-major-appliance-sales/
China’s Booming Home Appliance Market Hits a Wall
https://www.eetimes.com/chinas-booming-home-appliance-market-hits-a-wall/
Tomi Engdahl says:
Genius gamer builds a Nintendo Switch from scratch to battle price gougers
https://www.inputmag.com/gaming/genius-gamer-builds-a-nintendo-switch-from-scratch-to-battle-price-gougers
Tomi Engdahl says:
Apple Plans Its Own Chips for Macs
Apple Inc plans to sell Mac computers with its own main processors by next year based on the chip designs currently used in its iPhones and iPads. The iPhone maker is working on three Mac processors based on the A14 processor in its next iPhone, suggesting the company will transition more of its Mac lineup away from current supplier Intel Corp, the report here added citing people familiar with
Apple plans to sell Macs with its own chips from 2021: Bloomberg
https://www.reuters.com/article/us-apple-semiconductors/apple-plans-to-sell-macs-with-its-own-chips-from-2021-bloomberg-idUSKCN2251T5
Apple Inc (AAPL.O) plans to sell Mac computers with its own main processors by next year based on the chip designs currently used in its iPhones and iPads, Bloomberg reported on Thursday.
The iPhone maker is working on three Mac processors based on the A14 processor in its next iPhone, suggesting the company will transition more of its Mac lineup away from current supplier Intel Corp (INTC.O), the report here added citing people familiar with the matter.
Apple started using Intel’s processors in 2006 and a year later all Mac computers featured its chips.
Apple’s Mac computers generated $7.16 billion in revenue in the last reported quarter
Apple was planning to use its own chips in Mac computers beginning as early as 2020, Bloomberg had reported here in April 2018.
Tomi Engdahl says:
Chip Companies Well-Prepared for COVID-19 Pandemic, SEMI EHS Working Group Survey Shows
A recent survey by a SEMI Environment, Health and Safety working group found that all responding companies had Business Continuity Plans (BCP) in place prior to …
https://blog.semi.org/technology-trends/chip-companies-well-prepared-for-covid-19-pandemic-semi-ehs-working-group-study-shows
Tomi Engdahl says:
Practical Processor Verification
https://semiengineering.com/practical-processor-verification/
When creating a new processor, how much verification is required? Setting the right ROI is important.
Tomi Engdahl says:
An old Android phone can be used as a system monitor to show how hard your gaming PC is working
By Darren Allan 7 days ago
https://www.techradar.com/uk/news/an-old-android-phone-can-be-used-as-a-system-monitor-to-show-how-hard-your-gaming-pc-is-working
Others have used a Raspberry Pi to the same ends