PCWorld reviews Intel's recently-released benchmarks claiming Apple's M1 isn't faster than their 11th gen Core i7-1185G7 processor, among other things. Here are the claims Intel makes (visit the article to read PCWorld's "take" on each claim): MacBook M1 is slower than Core i7: Intel says in the WebXPRT 3 test, using the same version of Chrome for both the Core i7 system as well as the Arm-native MacBook, Intel takes the lead. The Intel chip was largely ahead in WebXPRT 3, and the x86 chip was nearly three times faster in finishing the photo enhancement test. Intel doesn't just use WebXPRT 3, though. It also shows the Core i7 pummeling the M1 in a PowerPoint-to-PDF export, and in multiple Excel macros by a factor of 2.3x. And yes, Intel used the Arm-native versions of Office for its tests.

Core i7 Crushes M1 in AI: For content creation tasks, Intel showed the Core i7 to be about 1.12x faster than the M1 in performing a 4K AVC-to-HEVC/H.265 file conversion. In this benchmark, they had the MacBook using the M1-native version of Handbrake. But the real destruction happens once you get to Topaz Lab's Gigapixel AI and Denoise AI, with the Intel Core chip crushing the M1 in AI-based noise removal and enlargement. Or maybe "crushing" is too nice a term, as it's more like the Core i7 outpaces the M1 by so much, the M1 wishes it had never been designed.

M1 doesn't support all the features: Intel also gives itself the lead in Adobe Premiere Pro, using the beta M1 native version in Auto Reframe, exporting to H.264 and H.265. They're decent wins, but come on, the code is still in beta for the Mac. That said, Intel points out that important features like Content Aware Fill are outright disabled on the beta version, and that's a concern. If the native version of Photoshop comes out, and there are critical features missing from it, that's a huge problem for Apple (and Adobe).

You can't be faster if you can't run it: For gaming, we see a bit of a back and forth between the Apple M1 and Core i7 in games that actually work on the MacBook. Intel doesn't let it end there, though, and decides to embarrass Apple further by showing the numerous games where the MacBook scores a 0 because game support just doesn't exist. Intel points out that "countless more" games "don't run on the M1," and then for good measure, it rushes Apple's bench with a list 10 more games you can't play on the M1 MacBook: Overwatch, Crysis Remastered, Halo MCC, Red Dead Redemption 2, PUBG, Monster, Hunter World, Doom Eternal, Microsoft Flight Simulator 2020, Apex Legends, and Rainbow Six Siege.

MacBook wouldn't win Evo certification: You know that fancy Intel Evo program that tries to improve laptop performance in key areas that annoy consumers? Well, Intel pretty much says that if Apple submitted the M1 MacBook to the same program that Asus, Dell, HP, Lenovo, MSI, Acer and others go through, it would be rejected. The reason? Intel says the M1 MacBook is too slow in doing things that anger consumers, such as "switch to Calendar" in Outlook, "start video conference Zoom" and "select picture menu" in PowerPoint.

Great battery life?: Perhaps the most shocking claim Intel showed deals with battery life. While performance tests can be cherry picked by those looking to prove an outcome, battery life usually can't be disputed. Apple's official claim gives the M1 MacBook up to 18 hours of battery life using Apple TV app to watch a 1080p video with the brightness set to "8 clicks from the bottom." Apple also claims up to 15 hours browsing 25 "popular" websites with the same "8 clicks" criteria. When Intel pitted a MacBook Air M1 against an Acer Swift 5 with a Core i7-1165G7, however, it found both basically dead even. The MacBook Air came in at 10 hours and 12 minutes, and the Acer Swift 5 lasted 10 hours and 6 minutes. The difference? Intel said it used Safari to watch a Netflix stream with tabs open with the screen set to a relatively bright 250 nits. On the Acer, Safari was subbed out for Chrome, but the brightness and Netflix remained the same. Intel did add that Apple's "8 clicks up" is about 125 nits of brightness on the MacBook Air which is pretty dim.

All kinds of things just don't work on the M1: Intel didn't just get into the performance of the M1. It also said it found the MacBook Pro had serious shortcomings, such as an inability to use more than one display with a Thunderbolt dock. And while the PC can use gaming headsets, eGPUs, a third-party finger print reader, Wacom Drawing tablet and Xbox Controller, Intel said it found the MacBook Pro simply doesn't work with eGPUs, and had multiple issues with other devices. That's just hardware incompatibility. Intel's rap battle with Apple also highlights issues with plug-ins for Ableton, Bitwig Studio, Avid Pro Tools, FL Studio, Motu and many others.

An anonymous reader quotes a report from Ars Technica: Homebrew now supports Apple Silicon natively, albeit not with every package. The volunteer Homebrew team made the announcement on the Homebrew blog alongside today's release. While the native support is not yet comprehensive, it bridges the gap significantly, and users can still run Terminal via Rosetta 2 to do what they can't yet while running natively on Apple Silicon. The Homebrew blog post says "we welcome your help" in providing bottles for all packages moving forward.

Here's the full bullet point on Apple Silicon in the Homebrew 3.0.0 release notes: "Apple Silicon is now officially supported for installations in /opt/homebrew. formulae.brew.sh formula pages indicate for which platforms bottles (binary packages) are provided and therefore whether they are supported by Homebrew. Homebrew doesn't (yet) provide bottles for all packages on Apple Silicon that we do on Intel x86_64 but we welcome your help in doing so. Rosetta 2 on Apple Silicon still provides support for Intel x86_64 in /usr/local."

Ryan Smith, writing at AnandTech: Following plans first unveiled last year during the launch of their DG1 GPU, Intel sends word this morning that the first Iris Xe video cards have finally begun shipping to OEMs. Based on the DG1 discrete GPU that's already being used in Intel's Iris Xe MAX laptop accelerators, the Iris Xe family of video cards are their desktop counterpart, implementing the GPU on a traditional video card. Overall, with specifications almost identical to Xe MAX, Intel is similarly positioning these cards for the entry-level market, where they are being released as an OEM-only part. As a quick refresher, the DG1 GPU is based on the same Xe-LP graphics architecture as Tiger Lake's integrated GPU. In fact, in broad terms the DG1 can be thought of as a nearly 1-to-1 discrete version of that iGPU, containing the same 96 EUs and 128-bit LPDDR4X memory interface as Tiger Lake itself. Consequently, while DG1 is a big first step for Intel -- marking the launch of their first discrete GPU of the modern era -- the company is planning very modestly for this generation of parts. The first DG1 GPUs were shipped in the fall as part of Intel's Iris Xe MAX graphics solution for laptops. At the time, Intel also indicated that a desktop card for OEMs would also be coming in 2021, and now, right on schedule, those desktop cards have begun shipping out. Further reading: Intel's Iris Xe DG1 Graphics Cards Not Compatible with AMD, Older Systems.

According to Bloomberg, Apple is working on a thinner and lighter version of the MacBook Air, the company's mass-market laptop. From the report: The new computer is planned to be released during the second half of this year at the earliest or in 2022. It will include Apple's MagSafe charging technology and a next-generation version of the company's in-house Mac processors. Apple has discussed making the laptop smaller by shrinking the border around the screen, which will remain 13-inches. The current model weighs 2.8 pounds and is just over half an inch at its thickest point.

The company considered building a larger version of the MacBook Air with a 15-inch screen, but Apple isn't moving forward with this for the next generation, said the people, who asked not to be identified discussing private matters. An Apple spokeswoman declined to comment. The new model will have a pair of USB 4 ports for connecting external devices. The new laptop is destined to be a higher-end version of the current MacBook Air, which is expected to remain in the company's lineup as an entry-level offering. Apple last updated the product in November with its own M1 Mac chip, replacing a processor from Intel Corp. Last Friday, Bloomberg reported on Apple's upgraded MacBook Pro laptops that are expected to be released later this year. They too will feature MagSafe charging, but unlike the MacBook Air, Apple's planning to bring back an SD card slot so users can insert memory cards from digital cameras. The Touch Bar is also going.

Samsung is considering spending more than $10 billion building its most advanced logic chipmaking plant in the U.S., a major investment it hopes will win more American clients and help it catch up with industry leader Taiwan Semiconductor Manufacturing Co. Bloomberg News: The world's largest memory chip and smartphone maker is in discussions to locate a facility in Austin, Texas, capable of fabricating chips as advanced as 3 nanometers in the future, people familiar with the matter said. Plans are preliminary and subject to change but for now the aim is to kick off construction this year, install major equipment from 2022, then begin operations as early as 2023, they said. While the investment amount could fluctuate, Samsung's plans would mean upwards of $10 billion to bankroll the project, one of the people said.

Samsung is taking advantage of a concerted U.S. government effort to counter China's rising economic prowess and lure back home some of the advanced manufacturing that over the past decades has gravitated toward Asia. The hope is that such production bases in the U.S. will galvanize local businesses and support American industry and chip design. Intel's troubles ramping up on technology and its potential reliance in the future on TSMC and Samsung for at least some of its chipmaking only underscored the extent to which Asian giants have forged ahead in recent years.

Intel said it was the victim of a hacker who stole financially sensitive information from its corporate website on Thursday, prompting the company to release its earnings statement ahead of schedule.ÂFrom a report: The US computer chipmaker believed an attacker had obtained advanced details about a strong earnings report it was due to publish after the stock market closed, said George Davis, chief financial officer. It published its formal earnings announcement upon discovering the problem, six minutes before the market closed. Intel's shares rose more than 6 per cent on Thursday, including almost 2 per cent in the final 15 minutes of trading. "An infographic was hacked off of our PR newsroom site," Mr Davis said. "We put [our earnings] out as soon as we were aware." He did not provide more details, but said that the leak was the result of an illicit action that had not involved any unintentional disclosure by the company itself. An Intel spokesperson added: "We were notified that our infographic was circulating outside the company. I do not believe it was published. We are continuing to investigate this matter."

We're following the state of play with Intel's new CEO, Pat Gelsinger, very closely. Even as an Intel employee for 30 years, rising to the rank of CTO, then taking 12 years away from the company, his arrival has been met with praise across the spectrum given his background and previous successes. He isn't even set to take his new role until February 15th, however his return is already causing a stir with Intel's current R&D teams. From a report: News in the last 24 hours, based on public statements, states that former Intel Senior Fellow Glenn Hinton, who lists being the lead architect of Intel's Nehalem CPU core in his list of achievements, is coming out of retirement to re-join the company. (The other lead architect of Nehalem are Ronak Singhal and Per Hammerlund - Ronak is still at Intel, working on next-gen processors, while Per has been at Apple for five years.)

Hinton is an old Intel hand, with 35 years of experience, leading microarchitecture development of Pentium 4, one of three senior architects of Intel's P6 processor design (which led to Pentium Pro, P2, P3), and ultimately one of the drivers to Intel's Core architecture which is still at the forefront of Intel's portfolio today. He also a lead microarchitect for Intel's i960 CA, the world's first super-scalar microprocessor. Hinton holds more than 90+ patents from 8 CPU designs from his endeavors. Hinton spent another 10+ years at Intel after Nehalem, but Nehalem is listed in many places as his primary public achievement at Intel. [...]

TSMC is to start making Intel's Core i3 on its 5nm process in 2H21 says TrendForce. From a report: Market analyst Trendforce reports that foundry TSMC is to start making Intel's Core i3 process later in the year on a 5nm process. This follows Intel's well documented problems with its leading edge process technology at 10nm and 7nm. The Core i3 move to a 5nm process is set to be followed by mid-range and high-end CPUs being produced for Intel by TSMC on a 3nm process in 2H22. TrendForce did not give a source for the information, simply referencing "investigations." Intel has long outsourced production significant amounts of its non-CPU chips to TSMC and UMC -- about 15 to 20 percent of its output, according to TrendForce. This is partly because it has often acquired fabless startups that had brought products to market using foundry. It was usually not worthwhile to re-engineer such products to Intel processes. It is also because Intel has wanted to focus on leading-edge specialist processes, although with less success in recent years. That 15 to 20 percent outsource was likely worth $10.5 billion to $14 billion in 2020, given Intel's annual revenue of $70bn.

Apple has plugged a hole that allowed users to sideload iOS and iPad applications to M1 Macs that were never intended to run on desktop. The server-side change ensures that only applications that app developers have flagged as optimized for Mac will run. From a report: Late last year, Apple launched its first Macs running on its own ARM-based custom CPU called the M1, as opposed to the Intel chips that have been used in Macs for several years. These new machines included the entry-level 13-inch MacBook Pro, the MacBook Air, and the low-end Mac mini. Since those machines now share an architecture with iPhones and iPads, which also have closely related ARM-based chips, it became possible to run iOS and iPadOS apps natively on Macs that were equipped with the M1 chip. Apple supported this by listing iPhone and iPad apps that passed an automated test on the Mac App Store, provided developers did not opt out of having the app listed. However, many developers did opt out for any number of reasons: because they did not feel the app provided a good user experience on laptops or desktops; because they offer preferred alternative ways to access services or content on Macs; because they don't have the time to support an additional platform; or any number of other reasons.

"The Trump administration notified Huawei suppliers, including chipmaker Intel, that it is revoking certain licenses to sell to the Chinese company and intends to reject dozens of other applications," Reuters reports, citing sources familiar with the matter: One of the sources said eight licenses were yanked from four companies. Japanese flash memory chip maker Kioxia Corp had at least one license revoked, two of the sources said. The company, formerly known as Toshiba Memory Corp, could not immediately be reached for comment... Companies that received the "intent to deny" notices have 20 days to respond, and the Commerce Department has 45 days to advise the companies of any change in a decision or it then becomes final. Companies would then have another 45 days to appeal...

Before the latest action, some 150 licenses were pending for $120 billion worth of goods and technology, which had been held up because various U.S. agencies could not agree on whether they should be granted, a person familiar with the matter said. Another $280 billion of licenses for goods and technology for Huawei still have not been dealt with, the source said, but now face a higher likelihood of denial... The United States made the latest decisions during a half dozen meetings starting on Jan. 4 with senior officials from the departments of Commerce, State, Defense and Energy, the source said. The officials developed detailed guidance with regard to which technologies were capable of 5G, and then applied that standard, the person said.

By doing that, the officials denied the vast majority of the roughly 150 disputed applications, and revoked the eight licenses to make those consistent with the new denials, the source said.

In addition to upgraded MacBook Pros, Bloomberg reports that Apple is also "planning the first redesign of its iMac all-in-one desktop computer since 2012," as it shifts away from Intel to its own silicon. From the report: The new models will slim down the thick black borders around the screen and do away with the sizable metal chin area in favor of a design similar to Apple's Pro Display XDR monitor. These iMacs will have a flat back, moving away from the curved rear of the current iMac. Apple is planning to launch two versions -- codenamed J456 and J457 -- to replace the existing 21.5-inch and 27-inch models later this year, the people said, asking not to be identified because the products are not yet announced. The new models will use next-generation versions of Apple's Mac processors like the upcoming 2021 MacBook Pros. The iMac redesign will be one of the biggest visual updates to any Apple product this year, according to people familiar with the company's roadmap.

Apple is also working on a pair of new Mac Pro desktop computers, its priciest Mac machines that don't come with a screen included, the people said. One version is a direct update to the current Mac Pro and will continue to use the same design as the version launched in 2019. Apple has discussed continuing to use Intel processors for that model rather than moving to its own chips. The second version, however, will use Apple's own processors and be less than half the size of the current Mac Pro. The design will feature a mostly aluminum exterior and could invoke nostalgia for the Power Mac G4 Cube, a short-lived smaller version of the Power Mac, an earlier iteration of the Mac Pro. Apple has also reportedly started development of a cheaper external monitor to sell alongside the Pro Display XDR. "The cheaper monitor would feature a screen geared more for consumer than professional use and wouldn't have the brightness and contrast ratio of the top-tier offering," reports Bloomberg.

Intel's new CEO, Pat Gelsinger, doesn't start his new role until February, but he's already prepping the company to take on Apple's M1 chips. From a report: The Oregonian, a local newspaper in Oregon where Intel maintains a large presence, reports that the chip maker held an all-hands company meeting yesterday, and Gelsinger attended. "We have to deliver better products to the PC ecosystem than any possible thing that a lifestyle company in Cupertino" makes, Gelsinger reportedly told Intel employees. "We have to be that good, in the future." Intel has been facing increased competition from both Apple and AMD recently. Apple announced its transition to its own silicon back in June, calling it a "historic day for the Mac." The transition has gone well, with M1-based Macs providing impressive performance and battery life compared to existing Intel-based Macs.

Apple is planning upgraded MacBook Pro laptops this year with much faster processors, updated displays and the return of its magnetic charger, Bloomberg reported Friday, citing a person with knowledge of the plans. From the report: The new laptops are planned to come in two screen sizes, a 14-inch model codenamed J314 and a 16-inch version internally dubbed J316. Both will use next-generation versions of Apple's in-house Mac processors, upgraded with more cores and enhanced graphics, the person said, asking not to be named as the products are not yet announced.

These devices will mark Apple's first high-end laptops to move away from Intel components. The company updated its base 13-inch MacBook Pro with its own M1 chip in November, to broadly positive reviews. Beyond the more powerful chips, Apple is also planning to step up the displays in its new MacBook Pros with brighter, higher-contrast panels, the person said. The new Macs will look similar to the current versions, albeit with minor design changes. Apple is aiming to launch the new MacBook Pros around the middle of the year. A major change to the new computers will be how they charge. Over the past five years, Apple has relied on USB-C ports for both power and data transfer on its laptops, making them compatible with other manufacturers' chargers. But the company is now bringing back MagSafe, the magnetic power adapter that means any accidental yanking of the power cable would simply detach it from the laptop rather than pull down the entire computer.

Intel CEO Bob Swan is set to step down effective Feb. 15. From a report: VMWare CEO Pat Gelsinger will take over the position, sources told CNBC. Intel's stock was up about 13% in premarket trading following the news. VMWare's stock was down nearly 5%. Swan was named CEO in January 2019 after serving as interim CEO for seven months. During Swan's tenure, Intel has suffered blows from competitors. Over the summer, Intel reported that its latest generation chips would be delayed while AMD's were already shipping inside laptops. Apple announced in the fall that it would use its own proprietary chips in its Mac computers, breaking a 15-year partnership with Intel for its chip supplies.

Advanced Micro Devices showed off some impressive Ryzen 5000 mobile processors today and teased the performance of its 3rd Gen Epyc server chips. From a report: Those chips are aimed at keeping AMD's performance lead over its rival Intel in the mobile and server markets. AMD CEO Lisa Su showed off the new chips in a keynote speech at CES 2021, the online-only tech trade show. AMD is launching its Ryzen 5000 Series mobile processors for gaming laptops and thin-and-light notebooks. These eight-core x86 chips are built with a 7-nanometer manufacturing process (where the circuits are 7 billionths of a meter apart). They are also based on the Zen3 design for processor cores, which can process instructions 19% faster per clock cycle than Zen2 cores.

The H-Series focuses on top performance in laptops for gamers and content creators, while the U-Series focuses on thin-and-light notebooks with great battery life. The chips have four to eight cores and they range in power consumption from 15 watts to 45 watts. AMD said the 5000 Series will be available in PCs in February, and we'll see more than 150 systems using it. That compares to 100 systems for the Ryzen 4000 Series and 70 for the Ryzen 3000.

MojoKid writes: At its virtual CES 2021 event today, Intel's EVP Gregory Bryant unveiled an array of new processors and technologies targeting virtually every market, from affordable Chromebooks to enthusiast-class gaming laptops and high-end desktops. Intel's 11th Gen Core vPro platform was announced, featuring new Intel Hardware Shield AI-enabled threat ransomware and crytpo-mining malware detection technology. In addition, the Intel Rocket Lake-S based Core i9-11900K 8-core CPU was revealed, offering up to a 19% improvement in IPC performance and the ability to out-pace AMD's Ryzen 9 5900X 12-core CPU in some workloads like gaming. Also, a new high-end hybrid processor, code-named Alder Lake was previewed. Alder Lake packs both high-performance cores and high-efficiency cores on a single product, for what Intel calls its "most power-scalable system-on-chip" ever. Alder Lake will also be manufactured using an enhanced version of 10nm SuperFin technology with improved power and thermal characteristics, and targets both desktop and mobile form factors when they arrive later this year.

Finally, Intel launched its new 11th Gen Core H-Series Tiger Lake H35 parts that will appear in high-performance laptops as thin as 16mm. At the top of the 11th Gen H-Series stack is the Intel Core i7-11375H Special Edition, a 35W quad-core processor (8-threads) that turbos up to 5GHz and supports PCI Express 4.0, and is targeted for ultraportable gaming notebooks. Intel is claiming single-threaded performance improvements in the neighborhood of 15% over previous-gen architectures and a greater than 40% improvement in multi-threaded workloads. Intel's Bryant also announced an 8-core mobile processor variant leveraging the same architecture as the 11th Gen H-Series that is slated to start shipping a bit later this quarter at 5GHz on multiple cores, with 20 lanes of PCIe Gen 4 connectivity.

Intel has talked with TSMC and Samsung about the Asian companies making some of its best chips, but the Silicon Valley pioneer is still holding out hope for last-minute improvements in its own production capabilities. Bloomberg reports: After successive delays in its chip fabrication processes, Santa Clara, California-based Intel has yet to make a final decision less than two weeks ahead of a scheduled announcement of its plans, according to people familiar with the deliberations. Any components that Intel might source from Taiwan wouldn't come to market until 2023 at the earliest and would be based on established manufacturing processes already in use by other TSMC customers, said the people, asking not to be identified because the plans are private.

Talks with Samsung, whose foundry capabilities trail TSMC's, are at a more preliminary stage, the people said. An Intel spokesperson referred to previous comments by Bob Swan, the company's chief executive officer. Swan has promised investors he'll set out his plans for outsourcing and get Intel's production technology back on track when the company reports earnings Jan. 21. [...] TSMC, the largest maker of semiconductors for other companies, is preparing to offer Intel chips manufactured using a 4-nanometer process, with initial testing using an older 5-nanometer process, according to the people. The company has said it will make test production of 4-nanometer chips available in the fourth quarter of 2021 and volume shipments the following year. The Taiwanese company expects to have a new facility in Baoshan operational by the end of this year, which can be converted to production for Intel if required, one of the people said. TSMC executives previously said the new Baoshan unit would house a research center with 8,000 engineers.

While Intel has outsourced production of lower-end chips before, it has kept the manufacturing of its best semiconductors in-house, considering it a competitive strength. Its engineers have historically tailored their designs to the company's manufacturing processes, making a shift to outsourcing of flagship products unthinkable in the past. As the provider of 80% of personal computer and server processors globally, Intel produces hundreds of millions of chips each year. That scale dictates that any potential supplier must create new capacity to accommodate Intel.

Dell is launching three new monitors next month, and all of them come with a dedicated Microsoft Teams button. The Verge reports: Dell claims it has created the "world's first video conferencing monitors certified for Microsoft Teams," after Microsoft started certifying displays, webcams, and headsets last year. Three monitors will be available next month, all offering quick access to Microsoft Teams. The button will let Microsoft Teams users quickly launch the app to make and receive video calls. Hands-free commands will also be supported through Cortana and the built-in microphone.

The Teams button is the main surprise with these displays, but Dell's trio of videoconferencing monitors also have some useful specs designed for an era where remote work and video calling is key. Each includes a 5-megapixel pop-up infrared camera, which supports facial recognition with Windows Hello. Dell also bundles a noise-canceling microphone and dual 5-watt integrated speakers. There's even a built-in mode to reduce blue light emissions. Dell is launching a 24-inch (FHD) version for $519.99, a 27-inch (QHD) model for $719.99, and a curved 34-inch (WQHD) variant for $1,149.99. All three will launch on February 16th. In other Dell-related news, the company has announced new versions of its Latitude laptops with Intel's new 11th Gen vPro chips and new features like an automated, integrated webcam shutter to physically block your camera when it's not in use.

An anonymous reader quotes a report from The Register: Linux creator Linus Torvalds has accused Intel of preventing widespread use of error-correcting memory and being "instrumental in killing the whole ECC industry with its horribly bad market segmentation." ECC stands for error-correcting code. ECC memory uses additional parity bits to verify that the data read from memory is the same as the data that was written. Without this check, memory is vulnerable to occasional corruption where a bit is flipped spontaneously, for example, by background radiation. Memory can also be attacked using a technique called Rowhammer, where rapid repeated reads of the same memory locations can cause adjacent locations to change their state. ECC memory solves these problems and has been available for over 50 years yet most personal computers do not use it. Cost is a factor but what riles Torvalds is that Intel has made ECC support a feature of its Xeon range, aimed at servers and high-end workstations, and does not support it in other ranges such as the Core series.

The topic came up in a discussion about AMD's new Zen 3 Ryzen 9 5000 series processors on the Real World Tech forum site. AMD has semi-official ECC support in most of its processors. "I don't really see AMD's unofficial ECC support being a big deal," said an unwary contributor. "ECC absolutely matters," retorted Torvalds. "Intel has been detrimental to the whole industry and to users because of their bad and misguided policies wrt ECC. Seriously. And if you don't believe me, then just look at multiple generations of rowhammer, where each time Intel and memory manufacturers bleated about how it's going to be fixed next time... And yes, that was -- again -- entirely about the misguided and arse-backwards policy of 'consumers don't need ECC', which made the market for ECC memory go away."

The accusation is significant particularly at a time when security issues are high on the agenda. The suggestion is that Intel's marketing decisions have held back adoption of a technology that makes users more secure -- though rowhammer is only one of many potential attack mechanisms -- as well as making PCs more stable. "The arguments against ECC were always complete and utter garbage. Now even the memory manufacturers are starting to do ECC internally because they finally owned up to the fact that they absolutely have to," said Torvalds. Torvalds said that Xeon prices deterred usage. "I used to look at the Xeon CPU's, and I could never really make the math work. The Intel math was basically that you get twice the CPU for five times the price. So for my personal workstations, I ended up using Intel consumer CPU's." Prices, he said, dropped last year "because of Ryzen and Threadripper... but it was a 'too little, much too late' situation." By way of mitigation, he added that "apart from their ECC stance I was perfectly happy with [Intel's] consumer offerings."

Long-time Slashdot reader UnknowingFool writes: AMD filed a patent on using chiplets for a GPU with hints on why it has waited this long to extend their CPU strategy to GPUs. The latency between chiplets poses more of a performance problem for GPUs, and AMD is attempting to solve the problem with a new interconnect called high bandwidth passive crosslink. This new interconnect will allow each GPU to more effectively communicate with each other and the CPU.
"With NVIDIA working on its own MCM design with Hopper architecture, it's about time that we left monolithic GPU designs in the past and enable truly exponential performance growth," argues Wccftech.

And Hot Hardware delves into the details, calling it a "hybrid CPU-FPGA design that could be enabled by Xilinx tech." While they often aren't as great as CPUs on their own, FPGAs can do a wonderful job accelerating specific tasks... [A]n FPGA in the hands of a capable engineer can offload a wide variety of tasks from a CPU and speed processes along. Intel has talked a big game about integrating Xeons with FPGAs over the last six years, but it hasn't resulted in a single product hitting its lineup. A new patent by AMD, though, could mean that the FPGA newcomer might be ready to make one of its own...

AMD made 20 claims in its patent application, but the gist is that a processor can include one or more execution units that can be programmed to handle different types of custom instruction sets. That's exactly what an FPGA does...

AMD has been working on different ways to speed up AI calculations for years. First the company announced and released the Radeon Impact series of AI accelerators, which were just big headless Radeon graphics processors with custom drivers. The company doubled down on that with the release of the MI60, its first 7-nm GPU ahead of the Radeon RX 5000 series launch, in 2018. A shift to focusing on AI via FPGAs after the Xilinx acquisition makes sense, and we're excited to see what the company comes up with.