Connect with us


Compliance with Moore’s Law is getting harder



There was a time, decades in fact, when all it took to make a better computer chip was smaller transistors and narrower interconnects. That time is long gone, and while transistors will continue to get smaller, it just doesn’t make sense to make them that way anymore. The only way to sustain the exponential pace of computing at present is through a scheme called joint system technology optimization, or STCO, the researchers say. ITF World 2023 last week in Antwerp, Belgium. It’s the ability to break chips down into their functional components, use optimal transistors and interconnect technologies for each function, and stitch them together to create a less power-hungry but better-functioning whole.

“This brings us to a new paradigm for CMOS,” says imek Research and Development Manager Marie Garcia Bardon. CMOS 2.0, as the Belgian nanotechnology research organization calls it, is a complex vision. But it may be the most practical way forward, and some of it is already evident in today’s most advanced chips.

How did we get here

In a sense, the semiconductor industry was corrupted for decades prior to 2005. Julien Ricart, Vice President of Research and Development, Imec. During this time, chemists and device physicists were able to routinely produce smaller, lower power, and faster transistors that could be used for every function on a chip, leading to ever-increasing computing power. But soon after that, the wheels began to move away from this scheme. Device engineers could have come up with some great new transistors, but these transistors didn’t allow for better, smaller circuits like SRAM and standard logic cells that make up the bulk of processors. In response, chip manufacturers began to break down the barriers between standard cell design and transistor development. The new scheme, called Design Technology Cooperative Optimization, or DTCO, has resulted in devices designed specifically to improve standard cells and memory.

But DTCO is not enough to keep computing running. The limitations of physics and economic realities conspired to put barriers to progress with the universal transistor. For example, physical limitations prevented the CMOS operating voltage from dropping below about 0.7 volts, slowing progress in power consumption. Anabela Veloso, chief engineer of Imec. Switching to multi-core processors helped solve this problem for a while. Meanwhile, I/O limitations meant that it became more and more necessary to integrate the functions of multiple chips into the processor. Thus, in addition to a system-on-a-chip (SoC) having multiple instances of processor cores, they also integrate network, memory, and often dedicated signal processing cores. Not only do these cores and features have different power and other needs, they also cannot be scaled down at the same rate. Even the CPU’s cache memory, SRAM, doesn’t shrink as fast as the CPU’s logic.

Joint optimization of system technologies

Getting things out of the dead end is not only a philosophical shift, but also a set of technologies. According to Rycart, STCO means considering the system-on-a-chip as a set of functions such as power supply, I/O, and cache memory. “When you start talking about features, you realize that SoC is not a homogeneous system, but just transistors and interconnects,” he says. “These are features optimized for different purposes.”

Ideally, you could build each feature using the most appropriate process technology for it. In practice, this basically means building each one on its own piece of silicon or chiplet. You then tie them together using technology such as advanced 3D stacking so that all functions work as if they were on the same piece of silicon.

Examples of this kind of thinking are already present in advanced AI processors and accelerators. The Intel Ponte Vecchio HPC accelerator (now called the Intel Data Center GPU Max) consists of 47 chiplets built using two different processes, each developed by both Intel and Taiwan Semiconductor Manufacturing Co. AMD is already using different technologies for the I/O chiplet. and compute chiplets in its processors, and recently the company began allocating SRAM for the high-level cache of compute chiplets.

Imec’s roadmap to CMOS 2.0 goes even further. The plan calls for further reductions in transistors, transferring power and possibly clock signals to CPU silicon, and increasingly tight integration of 3D chips. “We can use these technologies to recognize different features, separate the SoC and re-integrate it to achieve maximum efficiency,” Reikert says.

Transistors will change shape in the coming decade, but so will the metal that connects them. Ultimately, transistors can be composite devices made up of 2D semiconductors instead of silicon. Power supply and other infrastructure can be located under the transistors.imek

Transistor scaling continued

Major chip manufacturers are already moving away from the FinFET transistors used in computers and smartphones over the last decade to a new architecture, nanosheet transistors. [see “The Nanosheet Transistor Is the Next (and Maybe Last) Step in Moore’s Law”]. Ultimately, two nanosheet transistors will be built on top of each other to form a complementary field-effect transistor, or CFET, which Velloso says “represents the maximum scaling of CMOS.” [see “3D-Stacked CMOS Takes Moore’s Law to New Heights”].

As these devices shrink and change shape, one of the main goals is to reduce the size of standard logical cells. This is usually measured by “track height” – basically the number of metal connecting lines that can fit in the cell. Advanced FinFET FETs and early nanosheet devices are six-track cells. A five-track upgrade may require an intermediate design called a fork leaf that squeezes devices more closely without necessarily making them smaller. The CFETs will then reduce the number of cells to four lanes, or possibly less.

Four multi-colored blocks with arrows between them indicating progress.Advanced transistors are already moving from the fin field-effect transistor (FinFET) architecture to nanosheets. The end goal is to place two devices on top of each other in a CFET configuration. The fork leaf can be an intermediate step along the way.imek

Chip makers will be able to create the thinner features needed for this advancement using next-generation ASML lithography for extreme ultraviolet, Imec says. This technology, called High Numerical Aperture EUV, is currently under development at ASML and Imec will be next in line. Increasing the numerical aperture (an optical term referring to the range of angles at which a system can collect light) results in more accurate images.

Rear power supply networks

The basic idea behind internal power supply networks is to remove all interconnects that carry power—as opposed to data signals—above the silicon surface and place them below it. This should provide less power loss because the power supply interconnects can be larger and less stable. It also frees up space above the transistor layer for signal-carrying interconnects, possibly resulting in more compact designs. [see “Next-Gen Chips Will Be Powered From Below”].

In the future, even more could be moved to the back side of the silicon. For example, so-called global interconnects—those that span (relatively) long distances for transmitting clocks and other signals—can be hidden under silicon. Or engineers could add active power delivery devices such as ESD protection diodes.

3D integration

There are several ways of 3D integration, but the most advanced to date are hybrid wafer-to-wafer and die-to-wafer connections. [see “3 Ways 3D Chip Tech Is Upending Computing”]. These two provide the highest interconnection density between two silicon dies. But this method requires two dies to be designed together so that their functions and connection points match, allowing them to operate as a single chip, says Ann Jourdain, chief technical officer. Imec R&D is on track to create millions of 3D joints per square millimeter in the near future.

Going to CMOS 2.0

CMOS 2.0 takes disaggregation and heterogeneous integration to the extreme. Depending on which technologies make sense for specific applications, this can lead to a 3D system that includes layers of embedded memory, I/O and power infrastructures, high density logic, high drive current logic, and huge amounts of cache memory.

Achieving this will require not only technology development, but tools and training to determine which technologies will actually improve the system. As Bardon points out, smartphones, servers, machine learning accelerators, augmented and virtual reality systems are all very different requirements and constraints. . What makes sense for one may be a dead end for another.

From articles on your site

Related articles online


Hiltzik: COVID strategy that turned out to be deadly in practice



They have degrees from some of the most elite universities in the world – Harvard, Stanford, Johns Hopkins, Oxford. They are invited to the highest political councils of the government. They became an integral part of television news programs and were constantly quoted by some of the country’s leading newspapers.

This is a group of academics and scientists who have promoted a discredited solution to the COVID pandemic by avoiding masks, school closures and even vaccines, all in the name of achieving the elusive goal of “herd immunity” which has led to what may have been hundreds of thousands of unnecessary American deaths. .

This statement “We want them to get infected” a heavily documented new book by Jonathan Howard, an NYU neuroscientist and veteran exposing the pseudoscience that is polluting our pandemic efforts.

In 2019, you would be considered a charlatan if you assumed that the best way to get rid of a virus is to spread the virus. But it became mainstream and influenced politicians at the highest level.

— Jonathan Howard, MD

Howard received his title from Paul Alexander, an epidemiologist in the Trump administration’s Department of Health and Human Services.

In July 2020, Alexander expressed his opinion on how to use the relative risks of COVID to discrete populations to achieve herd immunity. The idea was that so many people would eventually become infected with the virus naturally, and therefore immune to further infection, that the virus would not be able to spread further.

“Infants, children, teens, young adults, young adults, middle-aged people with no medical conditions, etc. have zero or little risk,” he told senior HHS officials. “So we use them to develop the herd… we want them to be infected.”

Alexander’s proposal was essentially a screed against self-isolation. This suited the Trump White House, which was looking for ways to deal with the economic turmoil caused by the virus. But he was wrong about the number of illnesses and deaths that could result from allowing the virus to run rampant among these supposedly low-risk groups, and wrong about the prospects for achieving herd immunity naturally.

We Want Them to Get Infected is perhaps the most horrific and infuriating book you’ll ever read about America’s response to a pandemic. This is also important reading.

The book is populated by charlatans, charlatans, and charlatans, as well as quite a few scholars of outstanding academic achievement, many of whom seem to have been seduced into the embrace of the right-wing echo chamber to promote unproven and debunked policies.

“It’s incredible that while doctors like me were working to treat COVID patients, begging people to stay at home and be safe,” Howard told me, “another group of doctors were working against us—well-known doctors who were willing were intentionally infecting unvaccinated young people with a promise that herd immunity will appear in a couple of months.”

They consistently downplayed the severity of the pandemic, but rarely, if ever, admitted that their optimistic forecasts of morbidity and mortality were consistently wrong.

There are a number of problems with the theory of herd immunity. First, immunity from COVID infection tends to wane over time rather than become permanent. In addition, infection with one variant of the virus does not necessarily confer immunity from other variants, of which there were many.

Another concern is that COVID can be a devastating disease for victims of any age. Allowing someone to become infected can lead to serious health problems.

What’s more, the prospect that COVID can be beaten by natural expansion of herd immunity has convinced many people not to bother with proven countermeasures, including social distancing, mask-wearing, and vaccinations.

Today, more than three years after the first appearance of COVID, the US still has not achieved herd immunity, although it is getting closer to the goal. according to Robert Wachter, chair of the department of medicine at the University of California at San Francisco. The trajectory of the disease was catastrophic – the US death toll is 1.13 million.hundreds of children have died and it is estimated that 245,000 children have lost one or both parents to COVID. V USA leads the world in cases of death from COVID; its death rate of 3.478 per million population is worse than that of the UK, Spain, France, the Nordic countries, Canada, and Israel.

Some herd immunity advocates have offered their buoyant predictions in a misguided, if not dishonest, attempt to reassure the American public. Scott Atlas, a senior fellow at the Hoover Institution at Stanford University, in March 2020 urged HHS officials to speak out against the lockdowns on the grounds that they “inciting irrational fear” a virus that he estimates will kill about 10,000 people. “The panic needs to stop,” Atlas wrote.

Atlas soon became Trump’s top adviser, promoting herd immunity in the White House over the objections of more experienced advisers like Dr. Jones. Deborah Birks.

Howard is particularly concerned about how the politicization of the pandemic has allowed fringe ideas to infiltrate public health policy.

“In 2019, you would be considered a charlatan if you assumed that the best way to get rid of the virus is to spread the virus,” he says. “But it became mainstream and influenced politicians at the highest level.”

In his book, Howard treats propagandists with the deepest contempt. “The Great Barrington Declaration” the Herd Immunity Manifesto, published in October 2020 and originally signed by Stanford epidemiologist Jay Bhattacharya; Martin Kuhldorf, then from Harvard; and Sunetra Gupta of Oxford. (Thousands of other academics and scientists would later add their signatures.)

The essence of the declaration was to counter blocking. Its solution was what its drafters called “targeted protection,” which meant giving “those at the lowest risk of death a normal life to build up immunity to the virus through natural infection, while better protecting those most at risk.” . are mostly retirees.

The declaration states that older people living at home should be separated from other family members, except when meeting them on the street, and “should deliver groceries and other essentials to the home.”

Targeted protection, the promoters wrote, would allow society to achieve herd immunity and return to normal life in three to six months.

As Howard documents, the declaration was nothing more than a libertarian fantasy. Perhaps this was not surprising, because one of its organizers was an arch-libertarian named Jeffrey Tucker.

To get an idea of ​​Tucker’s worldview, consider a 2016 article titled “Let the kids work.” There he ridiculed the Washington Post for publishing photo gallery of working children 100 years ago, including miners and sweatshops as young as 10.

Tucker replied that these kids were “working in the adult world, surrounded by cool, fussy things and new technologies. They are on the streets, in factories, in mines, with adults and with peers, they study and do. They are valued for what they do, that is, they are valued as people … Whatever you say about it, it’s an exciting life.

At least a better life than “forced into government tanks for a whole decade” – that is, go to school.

The authors of the declaration, writes Howard, never specified how to achieve their goals. Delivering food and supplies to millions of housebound seniors? In an interview with the Hoover Institution, Bhattacharya said, “We could offer free DoorDash to the elderly.”

As Howard observes, Bhattacharya was surprisingly optimistic about “creating a program to deliver fresh food to tens of millions of elderly people for several consecutive months across the country.”

Hand-waving like this solved the problems of multi-generational households that house millions of vulnerable older people. The authors of the declaration wrote that elderly family members “may temporarily live with an older friend or sibling, with whom they can self-isolate together in the midst of community spread of infection.” As a last resort, vacant hotel rooms can be used for temporary housing.”

Of course, hermetically isolating tens of millions of “invulnerable” people from tens of millions of vulnerable people in a few weeks would be “the single greatest logistical challenge that mankind has ever faced,” notes Howard. “Nowhere in the world has targeted protection been used to achieve herd immunity in three to six months, as promised by the Great Barrington Declaration.”

The Declaration did indeed promote complacency. Its compilers, says Howard, were “people with no real responsibility for anything, which made the impossible seem very simple. The task of delivering food to the homes of the elderly was left to health authorities that were understaffed, overburdened and underfunded.”

What may be the most unforgivable element of the herd immunity movement is that children can be used as shields for the rest of the population. Its proponents advised against vaccinating young children on the grounds that their susceptibility to the virus was minimal or even non-existent so they could safely acquire immunity naturally – and possibly as Vinay Prasad of UC San Francisco impliesprovide increased immunity for adults in their families.

However, while children tend to experience fewer symptoms when they are infected, they are not immune. According to the Centers for Disease Control and Prevention, over 1600 American children under the age of 18 died from COVID during the pandemic.

In any case, death is not the only serious outcome from COVID. CDC says more than 14,000 children were hospitalized for COVID during a pandemic. Countless children may suffer long-term COVID or other lifelong manifestations of the disease. Howard said doctors advise deliberately infecting children with COVID when a vaccine is available, especially if the goal is to protect adults. Hey right.

In a science-driven world, the proponents of the failed herd immunity theory would have long lost their credibility and their public soap box.

The opposite happened. Bhattacharya and Kulldorf still have their platforms (Kuldorf is now affiliated with right-wing Hillsdale College). Both were nominated in December by Florida Gov. Ron DeSantis for president. “Committee on Public Health Integrity” accused of questioning federal public health policy.

Meanwhile, Scott Atlas was scheduled to give the commencement address at New College of Florida, the once-famous liberal arts institution that DeSantis has turned into a right-wing pedagogy paradise. He was met with exclamations from the audience of high school graduateshowever, indicates that America’s youth may not be as easily deceived as their parents.

At the moment, right-wing anti-science ideology seems to be on the rise. The COVID vaccine agitation is metastasizing into an opposition movement against all childhood vaccinations, a trend that threatens to spark a surge in other vaccine-preventable diseases such as measles and polio.

“The anti-vaccine movement saw an opportunity to sow doubt,” Howard told me. “Getting rid of all school vaccination regulations has always been the Holy Grail for them.”

Howard’s book is a warning. We may be on the brink of a public health disaster because the proponents of the failed theory that COVID can be fought with “natural immunity” without vaccines have been able to take on the mantle of truth-tellers. But it’s not.

Continue Reading


Deep sea mining hotspot teems with mysterious animals



Thousands of “otherworldly” animals live under the ocean in an area that can mine metals.

Continue Reading


How can minus 40 Fahrenheit equal minus 40 Celsius?



If you live in the US you probably use Fahrenheit, but most other places in the world use Celsius. Both were invented in the early to mid-18th century by European scientists, a Dutch physicist and inventor. Daniel Fahrenheit and Swedish astronomer Various centigrade.

The two scales differ in several ways. First, the freezing and boiling points of water are different. In Fahrenheit, water freezes at 32 degrees and boils at 212 degrees, while in Celsius, water freezes at 0 degrees and boils at 100 degrees. Since there are 180 degrees between freezing and boiling in Fahrenheit and only 100 degrees in Celsius, each degree is 1.8 times more Celsius than Fahrenheit.

Here’s how to easily convert between the two scales:

It is important to know that for every degree Celsius there is 1.8 degrees Fahrenheit (or 9/5 of a degree as a fraction).

Since the two scales start with different numbers, zero and 32, the conversion between them involves adding or subtracting 32.

To convert Celsius to Fahrenheit, multiply the Celsius temperature by 1.8 and then add 32. You can use this formula:

F = (1.8 x C) + 32

If you want to convert Fahrenheit to Celsius, subtract 32 from the Fahrenheit temperature and then divide the result by 1.8 using the following formula:

C \u003d (F – 32) ÷ 1.8

So (1.8 x -40°C) + 32 = -40°F.

This works the other way too. (-40F − 32) ÷ 1.8 = -40C

So why is this? Both scales must converge somewhere, and minus 40 degrees is just right. If you want to know the mathematical reason, keep reading.

Continue Reading


Copyright © 2023 Culture Belle Media.