Step into a world where the focus is keenly set on New Intel Chips 2022. Within the confines of this article, a tapestry of references to New Intel Chips 2022 awaits your exploration. If your pursuit involves unraveling the depths of New Intel Chips 2022, you've arrived at the perfect destination.
Our narrative unfolds with a wealth of insights surrounding New Intel Chips 2022. This is not just a standard article; it's a curated journey into the facets and intricacies of New Intel Chips 2022. Whether you're thirsting for comprehensive knowledge or just a glimpse into the universe of New Intel Chips 2022, this promises to be an enriching experience.
The spotlight is firmly on New Intel Chips 2022, and as you navigate through the text on these digital pages, you'll discover an extensive array of information centered around New Intel Chips 2022. This is more than mere information; it's an invitation to immerse yourself in the enthralling world of New Intel Chips 2022.
So, if you're eager to satisfy your curiosity about New Intel Chips 2022, your journey commences here. Let's embark together on a captivating odyssey through the myriad dimensions of New Intel Chips 2022.
New Intel chips want to make your laptop connect, work and play better
New Intel chips want to make your laptop connect, work and play better
Intel is adding six new laptop CPUs to its lineup, the company announced at this year's IFA technology trade show in Germany. Three new U-series and three new Y-series processors are joining Intel's 8th generation of Core i-series parts. Previously, these chips were known by the code names Whiskey Lake and Amber Lake.
These are for laptops, tablets and two-in-one devices that range from mainstream clamshells to thin, fanless designs, and include support for integrated gigabit Wi-Fi, which Intel says is a first for this class of consumer PC. Those system will be labeled as "Optimized for Connectivity."
Intel
The six new CPUs are in the Core-m and Core-i families and include:
1.1GHz Core m3-8100Y
1.3GHz Core i5-8200Y
1.5GHz Core i7-8500Y
2.1GHz Core i3-8145U
1.6GHz Core i5-8265U
1.8GHz Core i7-8565U
In introducing the new CPUs, Intel says that its goal was to address pain points like low-battery anxiety. By encouraging further power efficiencies, newer laptops can hit up to 16 hours of battery life. The U and Y-series chips may also provide enough of a performance boost to convince people holding into older computers to upgrade. Intel claims there are a large number of users still working on PCs more than five years old, and that the latest 8th-gen CPUs can usually double the performance of that 5-year-old system.
Intel is also focused on gaming, which is somewhat surprising, as the integrated graphics chips in Intel platforms have not kept up with PC gaming needs over the years. For most gamers, low-cost laptops and desktop with Nvidia graphics hardware for as little as $799 have filled some of that performance gap.
But Intel says it wants to support casual gaming, which the company defines as games like World of Warcraft or World of Tanks. To help buyers figure out what games they can play, and what settings to use, check out Intel's existing gameplay.intel.com website, which provides a list of compatible games, and suggested resolution and detail level settings based on your exact hardware, including updated setting for these new CPUs.
Intel
More importantly, although these new CPUs don't offer any new integrated graphics hardware, Intel says the overall efficiency of these chips, especially compared to the laptops of a few years ago, will offer smoother pen and inking performance and near-real-time 4K video rendering.
Look for laptops and tablets with these latest 8th-gen Intel CPUs to be announced at IFA and shipping soon after.
Laptops with the best battery life : See the top 25 laptops and 2-in-1 PCs with the longest battery life.
Great games for your non-gaming laptop : No GPU? No problem. The best games to sneak onto your work laptop.
Intel's Alder Lake chip could speed PCs by 30% while saving battery power
Intel's Alder Lake chip could speed PCs by 30% while saving battery power
Intel's new Alder Lake processor could boost the performance of personal computers by as much as 30% while delivering longer battery life, a breakthrough the chipmaker credits to a hybrid design that marries modules for top speed with others for efficient operation. It's an approach that's been used for years in smartphones.
At its Architecture Day event on Thursday, Intel said Alder Lake chips will come in three broad classes to power mainstream laptops, ultralight laptops and beefier desktop PCs. The three classes will be modeled on smartphone chips that combine high-performance computing cores for demanding jobs with smaller efficient cores that don't sap as much energy.
The number of performance and efficiency cores differ for each variety, but the fastest model will have eight of each. Intel's diagrams showed mobile Alder Lake chips combining six performance cores with eight efficiency cores and ultramobile chips combining two performance cores with eight efficiency cores. The Alder Lake family of chips will be available in PCs this fall.
The power and efficiency boost over today's 11th-gen Core models code-named Tiger Lake are a crucial selling point now that we have grown to expect laptops that can run for an entire day unplugged. Intel already competes with Apple and its M1 Mac processor that delivers improved power and battery life.
"We want [to] combine the best of both best of both in one system," Raja Koduri, who runs Intel's Accelerated Computing Systems & Graphics group, said in an advance briefing. Alder Lake's hybrid approach will help chart Intel's course for the next decade, Intel said.
Alder Lake's hybrid architecture is a variation of the Big.Little design that chip designer Arm rolled out a decade ago and now dominates smartphones. Apple employs the same approach with the M1 chip, which began powering Macs in 2020 and likely will be upgraded for more powerful Macs expected this year.
Alder Lake marks an important moment of unification and simplification for Intel's chip product line. Intel has been juggling a hodgepodge of new and old designs as it struggled with delays to its manufacturing process. The company fell behind its competitors as it wrangled with the problems, with AMD chip designs gaining market share, Apple ejecting Intel chips from its Macs and Asian giants Taiwan Semiconductor Manufacturing Co. (TSMC) and Samsung leapfrogging Intel in chip manufacturing.
Now all Intel's PC chips will be built with a manufacturing process called Intel 7. The same goes for Alder Lake's big sibling, Sapphire Rapids, a server chip due to arrive in the first half of 2022. Reducing the complexity should lower Intel's costs.
The chips were designed before Pat Gelsinger rejoined Intel earlier this year as chief executive. But they're an important part of his attempt to reclaim Intel's lost chipmaking leadership. He's also overseeing Intel's launch of a foundry business that builds chips for other companies, including rival Qualcomm, and hiring rival TSMC to build parts of its own chips.
"Intel is getting back to the Intel of old," said Tirias Research analyst Kevin Krewell, with the company rebuilding its past technical and operational acumen.
How much faster and more efficient will Alder Lake be?
Intel has been cagey with details but released some measurements of Alder Lake.
An average of speed tests shows Alder Lake's performance cores, known by the code name Golden Cove, demonstrates a 19% boost over today's Tiger Lake chips while running at the same clock speed, said Adi Yoaz, director of the Intel Core architecture. "This is our largest architectural shift in over a decade," he said.
Intel Alder Lake processors likely will all come with eight high-efficiency cores, shown in blue, for lower priority tasks and battery-saving operations. The number of high-performance cores, shown in purple, will range from eight on beefy desktop processors to two on ultramobile devices.
Intel
The new manufacturing process should add another 10% to 15% through hardware refinements that improve attributes like clock speed. Together, that could mean a 30% or more boost in top speeds, a big step up from past annual improvements typically less than 10%.
For preserving battery life while running a set of lower priority tasks, the Gracemont efficient core's design improves Intel's over earlier Skylake design, which is still widely used even though it's six years old. While juggling four tasks at once, efficiency cores require a fifth the power of Skylake cores, said Intel chip architect Stephen Robinson.
Using a technology called Thread Director, Alder Lake will determine whether computing jobs should be assigned to performance or efficiency cores or be shuffled around as new tasks arrive. Thread Director requires support built into Microsoft's upcoming Windows 11, but the current Windows 10 will be able to tap into some of its multicore features, said Rajshree Chabukswar, an engineer on the Alder Lake team.
Intel buying chips from rival TSMC
At Architecture Day, Intel revealed that TSMC, the foundry that builds Apple iPhone and Mac chips, will build Alchemist, the first member of Intel's new Arc family of standalone graphics chips. The chip is due early next year.
TSMC will also build Intel-designed graphics chips core to Ponte Vecchio, a massive processor package with high-speed links between many different chip elements. Ponte Vecchio will be the main brains of the Energy Department's $500 million Aurora supercomputer at Argonne National Laboratory. (Slow development of the processor has delayed the arrival of that machine.)
Expect Ponte Vecchio's array of packaging technologies -- an Intel manufacturing advantage -- to trickle down to mainstream products, said Real World Technologies analyst David Kanter.
"Ponte Vecchio is the Lamborghini of the chip world. Will this become the Lexus and then become the Toyota?" he asked. "The answer is yes."
Other Alder Lake chip specs
Alder Lake processors will come in varieties consuming as little as 9 watts for ultramobile devices and 125 watts for the beefy desktops used for the most demanding tasks like gaming and video editing.
The new chip is faster thanks to a variety of improvements in how it fetches instructions to execute, caches data instructions in high-speed memory, predicts the instructions it expects to run and recovers from mistakes in those predictions.
And it gets new instructions to execute artificial intelligence tasks with a technology called Advanced Matrix Extensions. AI is an immensely important new type of computer work, and many chip designers -- including Apple, Google, Qualcomm, Samsung, Arm and many startups -- are building dedicated AI acceleration electronics into chips.
Also built into Alder Lake are controllers to handle faster new memory standards, DDR5 (Double Data Rate 5) memory and, for mobile devices, LPDDR5 (Low Power Double Data Rate 5). Faster memory helps keep the processor fed with data so it doesn't have to spend as much time idling.
For connecting to the other devices like network controllers and graphics cards, Alder Lake chips debut support for the new fifth-generation PCI Express. That can double data transfer speed compared to PCIe 4.
Apple m1 chip performance apple m1 chip cores apple m1 processor vs intel what is the apple m1 processor apple m1 processor apple m1 processor review apple s m1 chip review apple smartphone apple smart keyboard
Apple's custom-built M1 processor and the new MacBook Airs, MacBook Pros and Mac Minis that use it are a problem for Intel. The divorce proceedings will last about two years as the prestigious customer gradually ejects Intel's chips from its personal computers.
But Intel isn't doomed.
The Santa Clara, California, company has some advantages and options in the PC market that insulate it from Apple's threat. Other PC makers aren't going to have as easy a time as Apple in moving past Intel. Intel is still the leader in higher-end chips more powerful than the M1. And it's got enough money on hand -- $18.25 billion in cash, equivalents and investments -- to let it spend its way to a better situation.
"There isn't much near-term threat to Intel's PC business beyond losing one sizable customer," said Linley Group analyst Linley Gwennap. That doesn't mean it's going to be easy for Intel, though.
Giving Apple grounds for divorce is the latest of the chipmaker's whiffs. Earlier achievements, like charting decades of steady chip industry progress with Moore's Law, pioneering PC technology standards and powering Google's data centers, have been overshadowed by newer flubs. That includes losing its manufacturing lead and failing to tap into the smartphone market. Intel ultimately sold its cellular chip business to Apple for $1 billion.
Though Macs account for only about 8.5% of the PC market, according to IDC, Apple remains one of the biggest and most influential tech companies. Its MacBook Air models led the trend to slim but useful laptops, its MacBook Pro models remain popular with programmers and the creative set, and Apple profits from selling premium machines costing hundreds of dollars more than most Windows PCs.
Losing Apple's business will sting. New Street Research analyst Pierre Ferragu estimated in a Wednesday report that 4% to 5% of Intel's revenue comes from Apple. But it's just one of the concerns Intel will need to address.
Intel said it's "relentlessly" focused on building leading chips. "We welcome competition because it makes us better," Intel said in a statement. "We believe that there is a lot of innovation that only Intel can do," including supplying chips that span the full price range of PCs and that can run older software still common in businesses.
It's also built its first samples of the 2021 Alder Lake PC chips and expects improvements in 2022 and beyond. "We're increasingly confident in the leadership our 2023 products will deliver," the company said.
Intel faces several challenges along the way, though.
The Qualcomm worry
One of the biggest concerns tied to the arrival of Apple's M1 is that it could embolden another Intel rival, Qualcomm, which already sells mobile-based processors for PCs.
The M1 is a member of the Arm family of processors that are used in every smartphone today. Qualcomm, a leading designer of those chips for Android phone makers, is pushing more-powerful versions of its Snapdragon chips for PCs, too, and several PC makers offer Windows laptops using them.
So far, though, Arm-based Windows laptops have shown lackluster performance and remain a rarity among customers. Arm PC makers have to prove better value and performance before more people adopt the machines, said CCS Insight analyst Wayne Lam.
Apple's transition to Arm-family M1 chips is also very different from Windows PC makers using Qualcomm chips. No PC maker is dumping Intel the way Apple is, so software makers don't need to worry as much about adapting their products for the new chip architecture. Though it might be nice to have, Qualcomm PC support isn't really essential.
The AMD threat
Intel is the dominant manufacturer of chips in the x86 family, which are the kind of processors you'd find in a normal laptop. But it's not the only x86 chipmaker.
"AMD is a greater threat in the near term," said Tirias Research's Kevin Krewell, who noted that PC makers aren't going to be quick to drop the industry standard family of x86 chips.
AMD has done well with high-end desktop processors, chiefly for gamers, and is making inroads in the server market, too. It's using Taiwan Semiconductor Manufacturing Corp. for manufacturing, taking advantage of its miniaturization progress to cram more circuitry onto new chips. Its new Zen 3 chip design offers a substantial speed boost.
In contrast, Intel, which manufactures its own chips, has struggled. It's only now moving in earnest from an earlier manufacturing technology with 14-nanometer features to a newer 10nm process after years of delays. Even next year's Rocket Lake chip for desktop computers will still be built with the 14nm process. (A nanometer is a billionth of a meter, and the smaller the measurement, the more transistors you can cram into a chip.)
AMD's Ryzen 5000 processor family, with up to 16 processing cores, challenges Intel in gaming PCs.
AMD
New manufacturing options
Intel is giving itself new options, including the ability to use other manufacturers like TSMC to build its chips. That's got risks, too, though, Gwennap said.
Moving some manufacturing to a partner makes it harder for Intel to justify the expense of trying to develop cutting-edge manufacturing, according to Gwennap. And the possibility that Intel could reclaim manufacturing once it fixes its problems could spook TSMC away from investing enough to meet Intel's massive demand.
Intel didn't comment on its manufacturing plan details. It said its integrated design and manufacturing approach helps competitiveness and in letting Intel assure customers it can supply the chips they need. "We've also been clear we will continue investing in leading process technology development," Intel said.
Apple, in contrast, has benefited from TSMC's steadily improved manufacturing. It's one reason it can fit a whopping 16 billion transistors onto its M1 chip, enough circuitry to power the main processor engines along with lots of extra abilities.
Apple's M1 starts small
Over and over during the new Mac launch event, Apple emphasized the performance per watt advantages of the M1. Translate that as being able to do useful work without draining a laptop battery fast.
Apple gets this advantage from the M1's lineage: the A series of processors that power iPhones. Smartphone chips have even stronger battery constraints than laptop chips. With the M1, a close relative of the iPhone 12's A14, Apple gets to add more transistor circuitry for more processing power and can run the chip at a higher clock speed than in phones, too.
Apple steadily increased A series chip performance for years, evolving the chip design and taking advantage of the prowess of TSMC, which manufactures the chips. Speed tests published by tech site Anandtech using the SPECint2006 benchmark show the A14 surpassing Intel's quad-core laptop chip, the 3GHz Core i7 1185G7 model that's a member of the new Tiger Lake processor family.
But the reality is that even Apple isn't ready to use the M1 in brawnier systems. The MacBook Air is all-in on M1, but Apple continues to rely on Intel for higher-powered 13-inch MacBook Pros. The 16-inch MacBook Pro, the iMac, the iMac Pro and the Mac Pro will continue to use Intel processors as Apple moves through a two-year transition to its own chips.
"It will get really interesting when Apple starts specifically optimizing its architecture for higher performance in a bigger thermal envelope and constant power for desktops," Techsponential analyst Avi Greengart said.
So yes, Intel has challenges. Apple's M1 is just the most obvious.
Intel Shows Off the Chip Tech That Will Power Your PC in 2025
Intel Shows Off the Chip Tech That Will Power Your PC in 2025
Intel on Thursday showed a silicon wafer studded with chips built with a manufacturing process that's set to arrive in 2025, a signal intended to reassure customers that the company's years of chip manufacturing difficulties are behind it.
"We remain on or ahead of schedule against the timelines that we laid out," Chief Executive Pat Gelsinger said of the company's plan to improve manufacturing processes. He showed off a gleaming wafer of memory chips built with the company's upcoming Intel 18A process, which overhauls the transistors at the heart of chip circuitry and the way power is delivered to them.
Intel is trying to dramatically accelerate manufacturing progress to meet a 2025 goal of reclaiming the chip performance lead it lost to Taiwan Semiconductor Manufacturing Co. (TSMC) and Samsung. If it succeeds, it'll mean PC chips progress faster after a half decade of lackluster performance improvements. And it could mean Intel becomes more relevant to your digital life by building chips inside your car, phone and gaming PC graphics card.
At the heart of the effort is moving through five new manufacturing processes in four years: Intel 7 in 2021 with the Alder Lake chips now powering PCs, Intel 4 in 2022, Intel 3 in 2023, Intel 20A in early 2024 and Intel 18A in late 2024 -- though the lag between manufacturing availability and product delivery means 18A chips won't arrive until 2025. Showing the wafer is a "proof point" that Intel is on track, Gelsinger said.
Gelsinger, a chip engineer who returned to Intel a year ago, brings tech cred to the CEO job, but it'll be tough for the company to claw its way back. Once a chip manufacturer falls behind the leading edge, as IBM and GlobalFoundries did in recent years, it's harder to justify the colossal investments needed to advance to the new technology.
Embodying Intel's difficulty is Apple's decision to eject Intel Core processors from its Macs in favor of its own M series chips built by TSMC. At the same time, AMD has been gaining market share, Nvidia has been profiting from gaming and AI, and Amazon has introduced its own server processors.
Gelsinger spoke at Intel's investor day, where he and other executives sought to convince often skeptical analysts that the company's enormous spending on new chipmaking equipment will pay off. That will come through premium products and external customers arriving to use its new foundry manufacturing capacity.
Intel 20A introduces two major changes to chip design, RibbonFET and PowerVia, and Intel 18A refines it for better performance. RibbonFET is Intel's take on a transistor technology called gate all around, in which the gate that governs whether a transistor is on or off is wrapped entirely around ribbon-like channels that carry the electrical current.
And PowerVia delivers electrical power to the underside of the transistor, freeing the top surface for more data link circuitry. Intel is playing catch-up with RibbonFET, but it's got a lead with PowerVia, which the industry calls backside power delivery.
Intel is pressing with another lead -- packaging technology that links different "chiplets" into one more powerful processor. The Sapphire Lake member of Intel's Xeon server family arriving this year employs one packaging variety, called EMIB, while the Meteor Lake PC chip arriving in 2023 employs another, called Foveros.
Intel expects to keep up with Moore's Law, which calls for a doubling in the number of transistors per processor every two years. That'll happen through smaller transistors and new packaging techniques combining multiple "chiplets" into one processor.
Intel
Intel built its first Meteor Lake prototypes in the final quarter of 2021 with the Intel 4 process and booted them up in PCs, said Ann Kelleher, the executive vice president who leads Intel's technology development division.
"This is one of the best lead product startups we have seen in the last four generations of technology," Kelleher said. "Over its lifetime, Meteor Lake will ship hundreds of millions of units, offering the clearest demonstration of leadership packaging technologies in high volume."
Packaging will play a role in future PC processors, including Arrow Lake in 2024, which will incorporate the first chiplets built with Intel 20A. After that comes Lunar Lake, which will use Intel 18A chiplets. Meteor Lake and Arrow Lake will use a new graphics chip architecture that Intel promises will be "a huge step forward," which is important given that graphics chips these days do a lot more than paint pixels on your screen -- for example AI and video image processing.
Kelleher also detailed a host of research and manufacturing changes to prevent the catastrophic problems Intel faced in recent years. For one thing, improvements are now modular, so a problem with one needn't derail others. For another, Intel is developing contingency plans for when problems do arrive. And it's paying more attention to the advice of chip equipment suppliers like ASML.
Nvidia grace cpu superchip nvidia ai graphics card nvidia grace cpu pdf nvidia grace cpu release date nvidia grace cpu pdf nvidia ai graphics card nvidia card for ai nvidia grace cpu specs grace hopper nvidia nvidia software nvidia support nvidia system monitor
Nvidia's Grace AI chip leaves Intel processors behind
Nvidia's Grace AI chip leaves Intel processors behind
Nvidia has a new chip in the works for boosting artificial intelligence and other high-performance computing work: Grace, a design slated to arrive in mammoth supercomputers in 2023. Instead of accelerating conventional Intel-powered servers, though, the design includes its own built-in Arm processors.
Nvidia's current brainiest chip, the A100, is typically yoked to Intel Xeon processors. Nvidia chips do the grunt work, but Intel chips oversee it. With Grace, named after pioneering programmer Grace Hopper, the company opted to embed several Arm Neoverse processor cores within the chip to speed up processing, said Paresh Kharya, an Nvidia senior director. The chip news arrived at Nvidia's GTC 2021 conference this week.
The new chip should let AI customers run computing tasks that are vastly more complex than is possible with today's chip designs, a step toward the general artificial intelligence that is the holy grail of today's machine learning research, said Cambrian AI Researach analyst Karl Freund in a blog post.
The design illustrates Nvidia's dramatic ascent -- and Intel's struggles. Even decades of dominance in technology don't guarantee success when the rules of computing are constantly being rewritten. Your laptop likely comes with an Intel chip, but an Nvidia chip was more likely responsible for important AI work like filtering spam, improving image quality or recognizing your voice when you call your bank.
Not so many years ago, Nvidia was just a component supplier, a designer of graphics chips called GPUs to boost PC performance. Intel's family of processors, or perhaps compatible rival AMD chips, shouldered most of the computing work. Intel, though, has struggled in recent years to keep pace with chip miniaturization and to capitalize on the exploding use of AI.
The result: Nvidia's market capitalization vaulted over Intel's, reaching $357 billion compared with Intel's $278 billion. Much of the growth has been propelled by the fact that GPUs also turned out to be pretty good at AI work, specifically the computationally intense training process that builds the models that later run in data centers, PCs and phones.
Also in the ascendant is Arm, which licenses the chip designs and technology that power every smartphone, new M1-based Apple Macs and the world's fastest supercomputer. Nvidia is seeking to acquire Arm for $40 billion, a move some rivals like Qualcomm object to. Grace's integrated Arm chips let Nvidia read data from memory many times faster than with current designs, the company said.
Nvidia's Selene machine, currently the world's fifth-fastest supercomputer, pairs A100 chips with AMD Epyc CPUs. A 2023 Grace-based machine called Alps at Switzerland's National Supercomputing Center should be seven times faster, Kharya said. The Los Alamos National Laboratory in the US also will buy a Grace-powered supercomputer.
Under new Chief Executive Pat Gelsinger, Intel is working to reclaim its manufacturing lead, planning to tap into others' manufacturing abilities while it works on miniaturizing its circuitry inscribing technology.
Intel is building AI abilities into its main processors while working on dedicated hardware, too. It folded its Nervana chips operation, but its Habana AI acceleration processors are still under active development.
One hot area for AI chips is autonomous vehicles, whose self-driving algorithms rely on processing in camera imagery and other sensor data. It's a core focus for Nvidia AI chip work, for example with its Orin chip scheduled to debut in 2022 vehicles.
Nvidia CEO Jensen Huang announced new processors for AI, graphics and supercomputing at the company's GTC event.
Screenshot by Stephen Shankland/CNET
At GTC, Nvidia announced a new chip called Atlan with quadruple the performance. It should arrive in 2025 vehicles, said Danny Shapiro, Nvidia's senior director of automotive work. Like Orin and Grace, Atlan relies on Arm cores, too.
Nvidia also announced a grander autonomous vehicle technology package called Hyperion 8. It combines two Orin processors with a host of sensors: eight exterior cameras, four exterior wider-angle fisheye cameras, three interior cameras, nine radar scanners and one lidar 3D scanner. The technology should arrive later in 2021.
Nvidia extended a partnership with Volvo, the companies said. Volvo plans to use Orin chips in its next-generation vehicles.
Intel has its own autonomous vehicle division, Mobileye. Tesla develops its own AI chips for its cars.
Nvidia's Grace AI chip leaves Intel processors behind
Nvidia's Grace AI chip leaves Intel processors behind
Nvidia has a new chip in the works for boosting artificial intelligence and other high-performance computing work: Grace, a design slated to arrive in mammoth supercomputers in 2023. Instead of accelerating conventional Intel-powered servers, though, the design includes its own built-in Arm processors.
Nvidia's current brainiest chip, the A100, is typically yoked to Intel Xeon processors. Nvidia chips do the grunt work, but Intel chips oversee it. With Grace, named after pioneering programmer Grace Hopper, the company opted to embed several Arm Neoverse processor cores within the chip to speed up processing, said Paresh Kharya, an Nvidia senior director. The chip news arrived at Nvidia's GTC 2021 conference this week.
The new chip should let AI customers run computing tasks that are vastly more complex than is possible with today's chip designs, a step toward the general artificial intelligence that is the holy grail of today's machine learning research, said Cambrian AI Researach analyst Karl Freund in a blog post.
The design illustrates Nvidia's dramatic ascent -- and Intel's struggles. Even decades of dominance in technology don't guarantee success when the rules of computing are constantly being rewritten. Your laptop likely comes with an Intel chip, but an Nvidia chip was more likely responsible for important AI work like filtering spam, improving image quality or recognizing your voice when you call your bank.
Not so many years ago, Nvidia was just a component supplier, a designer of graphics chips called GPUs to boost PC performance. Intel's family of processors, or perhaps compatible rival AMD chips, shouldered most of the computing work. Intel, though, has struggled in recent years to keep pace with chip miniaturization and to capitalize on the exploding use of AI.
The result: Nvidia's market capitalization vaulted over Intel's, reaching $357 billion compared with Intel's $278 billion. Much of the growth has been propelled by the fact that GPUs also turned out to be pretty good at AI work, specifically the computationally intense training process that builds the models that later run in data centers, PCs and phones.
Also in the ascendant is Arm, which licenses the chip designs and technology that power every smartphone, new M1-based Apple Macs and the world's fastest supercomputer. Nvidia is seeking to acquire Arm for $40 billion, a move some rivals like Qualcomm object to. Grace's integrated Arm chips let Nvidia read data from memory many times faster than with current designs, the company said.
Nvidia's Selene machine, currently the world's fifth-fastest supercomputer, pairs A100 chips with AMD Epyc CPUs. A 2023 Grace-based machine called Alps at Switzerland's National Supercomputing Center should be seven times faster, Kharya said. The Los Alamos National Laboratory in the US also will buy a Grace-powered supercomputer.
Under new Chief Executive Pat Gelsinger, Intel is working to reclaim its manufacturing lead, planning to tap into others' manufacturing abilities while it works on miniaturizing its circuitry inscribing technology.
Intel is building AI abilities into its main processors while working on dedicated hardware, too. It folded its Nervana chips operation, but its Habana AI acceleration processors are still under active development.
One hot area for AI chips is autonomous vehicles, whose self-driving algorithms rely on processing in camera imagery and other sensor data. It's a core focus for Nvidia AI chip work, for example with its Orin chip scheduled to debut in 2022 vehicles.
Nvidia CEO Jensen Huang announced new processors for AI, graphics and supercomputing at the company's GTC event.
Screenshot by Stephen Shankland/CNET
At GTC, Nvidia announced a new chip called Atlan with quadruple the performance. It should arrive in 2025 vehicles, said Danny Shapiro, Nvidia's senior director of automotive work. Like Orin and Grace, Atlan relies on Arm cores, too.
Nvidia also announced a grander autonomous vehicle technology package called Hyperion 8. It combines two Orin processors with a host of sensors: eight exterior cameras, four exterior wider-angle fisheye cameras, three interior cameras, nine radar scanners and one lidar 3D scanner. The technology should arrive later in 2021.
Nvidia extended a partnership with Volvo, the companies said. Volvo plans to use Orin chips in its next-generation vehicles.
Intel has its own autonomous vehicle division, Mobileye. Tesla develops its own AI chips for its cars.
Intel s loihi 2 speeds effort to make neuromorphic chips technology intel s loihi 2 speeds effort to make neuromorphic definition intel s loihi 2 speeds effort to make neuromorphic camera intel s loihi 2 speeds effort to calendar intel s loihi 2 speeds effort to equal crossword intel s loihi volcano intel s loihi shrimp intel s loihi surfacing intel s loihi definition intel stock intel software intel ssd intel security
Intel's Loihi 2 speeds effort to make neuromorphic chips like human brains
Intel's Loihi 2 speeds effort to make neuromorphic chips like human brains
Intel unveiled its Loihi 2 chip on Thursday, the second generation of a processor family that marries conventional electronics with the architecture of human brains to try to inject some new progress into the computing industry. On top of that, the chip also helps Intel advance its own manufacturing technology.
Loihi 2, an example of a technology called neuromorphic computing, is about 10 times faster than its predecessor, according to Intel. The speed improvement is the result of an eightfold increase in the number of digital neurons, a chip equivalent to human brain cells that mimic the way the brains handle information. The chip also can be programmed better to help researchers tackle more computing tasks.
The chip is built with a preproduction version of the Intel 4 manufacturing process, too, an advanced method Intel plans to use to build mainstream Intel chips arriving in 2023. The Intel 4 process can etch electronics more densely on a chip, a crucial advantage for Intel's need to pack a million digital neurons on a chip measuring 30 square millimeters.
Loihi chips are particularly good at rapidly spotting sensory input like gestures, sounds and even smells, says Mike Davies, leader of the Intel Labs group that developed Loihi. Some experiments have focused on artificial skin that could give robots a better sense of touch. "We can detect slippage if a robot hand is picking up a cup," Davies said.
Neuromorphic computing differs from artificial intelligence, a revolutionary computer technology based more loosely on how brains learn and respond, because it focuses more on the physical characteristics of human gray matter.
It differs from conventional chips in profound ways. For example, Loihi 2 stores data in tiny amounts spread across its mesh of neurons, not in a big bank of traditional computer memory, and it doesn't have a central clock ticking to synchronize computing steps on the chip.
You won't see Loihi 2 in your phone or laptop. Instead, it's geared for researchers at automakers, national labs and universities. Germany's Deutsche Bahn railway network is testing how well it can optimize train schedules. The processor is geared for tasks such as processing sound or detecting hand gestures, but with vastly lower power consumption, Davies said.
Neuromorphic computing
Low power use is a characteristic of biological gray matter, too. Human brains are made of about 80 billion cells called neurons, connected into elaborate electrical signaling networks. When enough input signals reach an individual neuron, it fires its own signal to other neurons. The topology of the connections and flow of signals lets us do everything from recognizing Abraham Lincoln to riding a bicycle. Learning is the process of establishing and reinforcing those connections.
Intel isn't the only one pursuing the idea. The Human Brain Project in Europe includes neuromorphic computing in its work. The way blood courses through the brain inspired IBM to power and cool chips with liquids in a flow battery. Samsung used IBM's neuromorphic TrueNorth chip to re-create vision.
Intel's chip is made of a million digital neurons that can be connected in any number of ways, a digital tabula rasa. Getting it to work requires configuring the proper connections between neurons. Actual processing occurs when input data reaches the chip, triggering a spike of activity that flows through the interconnected neurons and eventually produces an output. Each neuron is connected to 100 others on average, though some may reach as many as 10,000.
This flowlike design means the chip requires very little power when idle and can process data very quickly on demand, Davies said.
Programming neuromorphic chips is a big challenge, Davies said. To try to make the process easier for researchers, Intel also released an open-source software framework called Lava.
Fewer but smarter neurons
A million neurons in one chip is far from the billions in a human brain, but Intel is effectively trying to make each neuron smarter than a biological brain cell. For example, in biological brains, electrical signals are either fully on or fully off. In Loihi chips, Intel can assign a different strength to each signal, increasing processing sophistication, Davies said.
The chip can be connected to others, too, for greater scale. One improvement over the first Loihi is better networking that shortens the communication pathways that link neurons.
"The brain achieves accuracy and reliability through tremendous redundancy," Davies said. "The hope is indeed we can solve some of the same problems in a more economical way."
Intel 4 manufacturing
The Intel 4 process is a major step for Intel, its first move to a chipmaking technology called extreme ultraviolet, or EUV. Chip circuitry is etched onto silicon wafers using patterns of light, and EUV enables finer patterns for smaller structures, crucial to miniaturization and other improvements. Problems in recent years meant Intel lost its manufacturing leadership to Samsung and Taiwan Semiconductor Manufacturing Co. (TSMC), and Intel 4 is a key part of Intel's effort to reclaim that leadership by 2025.
Mainstream manufacturing with Intel 4 won't begin until the second half of 2022, with the resulting chips arriving in 2023, Intel has said. But Loihi 2 gives the company a chance to debut the technology while it's still in a preproduction phase.
"We have small quantities in the lab now," Davies said.
Intel made thousands of first generation Loihi chips, Intel said, and Loihi 2 likely will involve a similarly small production run. Manufacturing at Intel's full scale of operations, with millions of processors, brings other challenges.
M2 macbook air in store buy m2 macbook air m2 macbook air in store buy m2 macbook air apple smart glasses apple sign in apple student discount canada
Apple's M2 MacBook Air Available for Preorder Starting July 8
Apple's M2 MacBook Air Available for Preorder Starting July 8
Apple's new MacBook Air with M2 chip will be available to order on Friday starting at 5 a.m. PT, the tech giant said on Wednesday. Orders are expected to start arriving at customers' locations on July 15.
The M2 MacBook Air, which was unveiled in June at Apple's Worldwide Developers Conference, features the company's next-gen Apple Silicon chip, a fanless body in four color options and a 13.6-inch Liquid Retina display.
Apple said its M2 chip brings improved speed and efficiency to the new MacBook Air. The M2 makes "intensive workloads like editing complex timelines in Final Cut Pro" nearly 40% faster than the previous generation and applying "filters and effects in apps like Adobe Photoshop" up to 20% faster, Apple said.
The M2 MacBook Air starts at $1,199 (£1,249, AU$1,899) with an eight-core CPU and eight-core GPU, 8GB of memory and a 256GB solid-state drive. There's also a $1,499 version with an eight-core CPU and 10-core GPU, 8GB of memory and a 512GB SSD.
The redesigned laptop will be available to order on Apple's website and in the Apple Store app. It's available in available in four colors: silver, space gray, starlight and midnight.
§
This story is part of WWDC 2022, CNET's complete coverage from and about Apple's annual developers conference.
Apple on Monday debuted the new M2 processor, a chip that improves core processing performance 18% over the M1 without hurting battery life in the company's new MacBook Air and 13-inch MacBook Pro laptops.
The 18% speed boost comes from the M2's redesigned central processing units. The processor has four fast CPU cores and four efficient cores, a hybrid approach drawn from the smartphone world. By redesigning the graphics processing units and increasing their count up to a maximum of 10 instead of eight for the M1, GPU performance is 35% faster. Overall, the new MacBook Air is 20% faster at Photoshop image editing and 38% faster at Final Cut Pro video editing, Apple said.
"We continue to have a relentless focus on power-efficient performance," Johny Srouji, Apple hardware team leader, said at the Worldwide Developers Conference.
Power efficiency is crucial to shrinking laptops since the biggest component is the battery. The new MacBook Airs take up 20% less volume but still have a long, 18-hour battery life, Apple said. The company also is using the M2 in a new 13-inch MacBook Pro.
Apple's M2 processor has large amounts of high-speed cache memory built onto the chip itself and up to 24MB of regular memory included in the chip package, two attributes that should boost performance over Apple's 2020-era M1.
Apple/Screenshots by Stephen Shankland/CNET
The M2 processor also has a significant memory boost, reaching up to 24GB instead of 16GB for the M1. Memory is important, especially as software gets bigger and laptops have years-long lifespans. M series chips build memory directly into the processor package for fast performance, but it's not upgradable.
Apple debuted the M1 at 2020's WWDC and began shipping it later that year in the earlier version of the MacBook Air. The M1, along with beefier successors called the M1 Pro, M1 Max and M1 Ultra, struck an effective balance between performance and battery life and earned strong reviews.
The M2 doubles down on the same balanced approach, offering updated processing cores that are variants of the chips at the heart of newer iPhones. The new chips continue the gradual ejection of Intel processors from the Mac family of personal computers and could enable the last Intel-powered member, the Mac Pro, to switch to Apple chips.
Designing processors is an expensive, difficult undertaking. But with the M series chips, Apple takes advantage of the A series chip design work it already does for its iPhones and iPads, then pays Taiwan Semiconductor Manufacturing Co. to build the chips on its advanced product lines.
The M2 is built on TSMC's 5nm (5 nanometer) manufacturing process, but it's an improved version to the one used for the M1. TSMC is working on a more advanced 3nm process that should let customers squeeze in somewhat more transistors, the core electronics elements that process data on a chip.
The M2 has 20 billion transistors, a 25% increase over the M1, Apple said.
One use of the new transistors is the increased GPU count. Another is an upgraded neural engine -- a chip block used to accelerate artificial intelligence workloads. The new 16-core neural engine can perform 15.8 trillion operations per second, Apple said, a 40% speed boost.
With its own chips, Apple gets more control over the technology foundation of its products -- a principle important to Chief Executive Tim Cook. That includes both the processor itself, with specific features like AI acceleration, video encoding, and security, and the software Apple writes to take advantage of those features.
Apple's M series and A series chips are members of the Arm processor family. UK-based Arm licenses designs that companies can customize to varying degrees. Arm chips from Qualcomm, Apple, MediaTek, Samsung, Google and others power just about every smartphone for sale.
The Apple M2 processor is significantly larger than the M1. That increases manufacturing costs. Apple raised prices for its M2-based MacBook Air laptops.
Apple/Screenshot by Stephen Shankland/CNET
Intel has struggled over most of the last decade with problems advancing its manufacturing. That stalled its progress while Apple, Qualcomm, AMD, Nvidia and other Intel rivals took advantage of TSMC's manufacturing progress.
Because Apple doesn't offer its chips to others, and because the majority of PCs use Intel processors, Intel is somewhat insulated from Apple's shift. Intel is working to modernize its manufacturing, spending tens of billions of dollars on new chipmaking fabs. Intel aims to reclaim its lead over rivals TSMC and Samsung in 2024.
Intel's newest PC processor, code-named Alder Lake, embraces the same mix of high-performance and high-efficiency CPU cores found in smartphone chips and Apple's M series chips. Future products are designed to improve GPU performance, in particular with Intel's renewed focus on high-end graphics that's designed to wean the company from reliance on AMD and Nvidia. That's important for one big market, gaming, where PCs with Intel and AMD processors are much more widely used than Macs.
Things that are not as they appear things are not as they appear things that are not as they seem things that are not as they appear things which are seen were not made things that should have never been invented things that are not well known about things that have not been discovered 6 things you didn t know would happen things that never happened
6 Things That Didn't Make an Appearance at the Apple Event
6 Things That Didn't Make an Appearance at the Apple Event
Apple introduced an array of new devices at its "Peek Performance" event on Tuesday: a 2022 5G iPad Air with an M1 processor, an upgraded iPhone SE with 5G (and the same A15 chip as the iPhone 13) and the desktop-and-monitor pairing of the Mac Studio and Mac Studio Display. The M1 Ultra, a new processor even more powerful than the M1 Max, also took the stage as a chip option available for the Mac Studio.
While the newly announced products aligned with a lot of the speculation preceding the event (here's what the 2022 iPhone SE rumors did and didn't get right), not every rumored device ended up making an appearance. Here are all the Apple products that didn't get launched during Tuesday's event -- and that we're still hoping to see in 2022.
Upgraded Mac Pro
The Mac Pro, a desktop tower targeted at professionals, is due for an upgrade. It's the only remaining Mac that hasn't yet made the leap from an Intel processor to one of Apple's own chips (Apple discontinued the 27-inch iMac Tuesday instead). And we're right at Apple's self-imposed two-year deadline to stop using third-party processors, so it's more than likely we'll see a new Mac Pro, maybe with a brand-new M1 Ultra chip. It's possible this desktop will show up at Apple's next event. Check out our full roundup of Mac Pro rumors.
Faster Mac Mini
The 2020 Mac Mini with Apple's M1 processor could see an update powered by the M1 Pro, M1 Max or even the newly announced M1 Ultra chip. Reliable tech analyst Mark Gurman predicted in February that a new Mac Mini would be one of the first Macs that Apple introduced this year. While the Mac Studio technically proved him wrong at Apple's Tuesday event, Gurman did say more Macs would follow in May or June. A 2022 Mac Mini could be one of them.
iPad Pro and entry-level iPad
The rumor mill speculated that the iPad Pro and entry-level iPad would get upgrades at Apple's March event, but Apple instead introduced the 2022 iPad Air. Considering Apple's typical product release timelines, the next possible arrival times for the iPad and iPad Pro are either in June or sometime in the fall. Speculation says a 2022 iPad Pro could include MagSafe and a battery upgrade. Here are the most up-to-date rumors about the next iPad Pro.
New MacBook Pro and MacBook Air
No MacBooks showed up at the Apple event, but new versions of the MacBook Pro and MacBook Air are still on the radar for this year. Rumors suggest that both the new Air and new Pro could come with a new M2 chip. Or maybe it'll be the M1 Ultra announced on Tuesday.
Other Apple products we could see in 2022
We're also expecting the iPhone 14 in the fall. Much less likely for this year is a foldable iPhone, which Apple has reportedly pushed back to 2025. Some new accessories could be on the horizon: AirPods Pro 2, an Apple Watch 8 with new health features or even the long-awaited Apple VR/AR headset. A couple of home entertainment devices are reportedly in the works as well. On the software front, look out for iOS 16 and MacOS 13, both of which could get a preview at the next Apple event. Speaking of which…
When is the next Apple event?
Now that Apple's spring event is over, we probably won't see another until June. That's Apple's annual Worldwide Developers Conference. WWDC is a goldmine of Apple announcements, and we'll be keeping our eyes out for any confirmation of the dates for WWDC 2022. In the meantime, here's everything we learned at WWDC 2021.
§
Despite hopes of Apple unveiling an updated 27-inch iMac, the company instead discontinued the 27-inch version of the desktop altogether. That change comes amid a day full of news out of Tuesday's Apple event, including new iPhone models and new, smaller desktop devices, which it seems are intended to replace the 27-inch iMac.
The 27-inch iMac is one of the last Macs that still run on Intel rather than Apple's own chips. Apple didn't immediately respond to a request for comment on why it discontinued the desktop.
Apple's most recent 27-inch iMac launched in 2020, and was priced at $1,799, $1,999 and $2,299. It added higher-end CPU and GPU choices over the previous iteration, also expanding RAM and storage and adding nano-texture screen coating.
You can check out CNET's live blog to see how Tuesday's Apple event unfolded, or go here for everything Apple announced.
