An anonymous reader quotes a report from ArsTechnica: We've said it before, and we'll say it again: USB-C is confusing. A USB-C port or cable can support a range of speeds, power capabilities, and other features, depending on the specification used. Today, USB-C can support various data transfer rates, from 0.48Gbps (USB 2.0) all the way to 40Gbps (USB4, Thunderbolt 3, and Thunderbolt 4). Things are only about to intensify, as today the USB Implementers Forum (USB-IF) published the USB4 Version 2.0 spec. It adds optional support for 80Gbps bidirectional bandwidth as well as the optional ability to send or receive data at up to 120Gbps. The USB-IF first gave us word of USB4 Version 2.0 in September, saying it would support a data transfer rate of up to 80Gbps in either direction (40Gbps per lane, four lanes total), thanks to a new physical layer architecture (PHY) based on PAM-3 signal encoding. For what it's worth, Intel also demoed Thunderbolt at 80Gbps but hasn't released an official spec yet.

USB4 Version 2.0 offers a nice potential bump over the original USB4 spec, which introduced optional support for 40Gbps operation. You just have to be sure to check the spec sheets to know what sort of performance you're getting. Once USB4 Version 2.0 products come out, you'll be able to hit 80Gbps with USB-C passive cables that currently operate at 40Gbps, but you'll have to buy a new cable if you want a longer, active 80Gbps. In a statement to CNET, USB-IF said they don't expect to see supporting products for the new spec for "at least 12 to 18 months."

The USB Implementers Forum also updated the USB Type-C Cable and Connector and USB Power Delivery specifications today to accommodate USB4 Version 2.0.

Over 290 MSI motherboards are reportedly affected by an insecure default UEFI Secure Boot setting that allows any operating system image to run regardless of whether it has a wrong or missing signature. From a report: This discovery comes from a Polish security researcher named Dawid Potocki, who claims that he did not receive a response despite his efforts to contact MSI and inform them about the issue. The issue, according to Potocki, impacts many Intel and AMD-based MSI motherboards that use a recent firmware version, affecting even brand-new MSI motherboard models.

Apple has announced new 14- and 16-inch MacBook Pros, featuring its latest M2 Pro and Max chips. From a report: The M2 Pro model will launch with a 12-core CPU, up to 19-core GPU, and up to 32GB of unified memory, while the M2 Max includes up to 38 cores of GPU power and support for up to 96GB of unified memory. The new 14-inch MacBook Pro with M2 Pro starts at $1,999, with the 16-inch model starting at $2,499. Both are available to order online today and will start shipping and appearing in Apple stores on January 24th.

Apple says the M2 Pro has double the amount of transistors the M2 shipped with and nearly 20 percent more than the M1 Pro. It also features 200GB/s of unified memory bandwidth, twice what's available on the regular M2. All of this power should result in better performance in apps like Adobe Photoshop and Xcode. Apple claims the MacBook Pro with M2 Pro "is able to process images in Adobe Photoshop up to 40 percent faster than with M1 Pro, and as much as 80 percent faster than MacBook Pro with an Intel Core i9 processor." The M2 Max chip has the same 12-core CPU as the M2 Pro, but much like the M1 Max, it really pushes the GPU power more. Apple claims the M2 Max is up to 30 percent faster than the M1 Max in graphics and can apparently "tackle graphics-intensive projects that competing systems can't even run." Chips aside, the latest MacBook Pro models now include Wi-Fi 6E3 and a "more advanced HDMI" (probably HDMI 2.1) that supports 8K displays up to 60Hz and 4K displays up to 240Hz.

The PC market is in rough waters, and it was for much of last year. Every PC maker except Apple saw year-over-year decline. Laptop sales are said to have suffered the most. From a report: This all made for a somewhat uncertain backdrop heading into CES 2023, the annual conference where tech companies show off the products they'll be releasing in 2023. Throughout the show, executives and representatives from various PC manufacturers acknowledged that the industry has a big task ahead of it this year: keep the laptop exciting. Some companies are trying to do that with goofy hardware things (such as Lenovo's dual-screen, dual-OLED, and touchpad-less Yoga Book 9i). But others are moving away from hardware -- and the raw power that hardware can provide -- and emphasizing quirkier software capabilities in this year's lineups. AMD revealed that some of its new chips will come with its first Ryzen AI engine, built on its XDNA architecture.
Intel's upcoming Meteor Lake chips will also bring AI capabilities.
HP announced new features for its Omen Gaming Hub, including integration with Nvidia's GeForce Now, and new remote management and insight services for IT. A consumer Windows PC, the Dragonfly Pro, was also unveiled, with its integration with HP's new "live concierge" service touted as a highlight. The report adds: And HP isn't alone in this conviction -- quite a few other manufacturers that had a big presence at CES this year emphasized showy software features that utilized camera tracking and AI, from Asus' and Acer's glasses-free 3D displays to Razer's soundbar that follows your head around to optimize your music. Even Lenovo's aforementioned dual-screen Yoga Book is a software offering in many ways; the form factor is hardly new, but the investments Lenovo has made in an impressive system of gesture control are what make it a viable product.

Initially mentioned during their Innovation 2022 opening keynote by Intel CEO Pat Gelsinger, Intel has unveiled its highly anticipated 6 GHz out-of-the-box processor, the Core i9-13900KS. The Core i9-13900KS has 24-cores (8P+16E) within its hybrid architecture design of performance and efficiency cores, with the exact fundamental specifications of the Core i9-13900K, but with an impressive P-core turbo of up to 6 GHz. From a report: Based on Intel's Raptor Lake-S desktop series, Intel claims that the Core i9-13900KS is the first desktop processor to reach 6 GHz out of the box without overclocking. Available from today, the Core i9-13900KS has a slightly higher base TDP of 150 W (versus 125 on the 13900K), 36 MB of Intel's L3 smart cache, and is pre-binned through a unique selection process to ensure the Core i9-13900KS's special edition status for their highest level of frequency of 6 GHz in a desktop chip out of the box, without the need to overclock manually.

The Core i9-13900KS has been a long-awaited entrant to Intel's Raptor Lake-S for desktop series, with previous reports from Intel during their Innovation 2022 keynote that a 6 GHz out-of-the-box processor was on the horizon for this year. As Intel highlights, the Core i9-13900KS represents a significant milestone for desktop PCs, with its 6 GHz out-of-the-box P-Core turbo frequency. This makes it one of the fastest desktop x86 processors, at least from the perspective that users don't need to overclock anything to attain these ridiculous core frequencies. From Intel's sneak peek video on YouTube published on Jan 10th, the Core i9-13900KS looks to have reached 6 GHz on two of the eight performance (P) cores, with a clock speed of up to 5.6 GHz on the remaining six cores, which is very impressive.

One of the adjustments Intel needed to make to power limitations to achieve these frequencies is somewhat hazy. Intel hasn't specified if the Core i9-13900KS is a special binned part, but from previous KS launches, this has been the case, and it's expected that it is still the case. The reports of Core i9-13900K chips being overclocked to 6 GHz at ambient are few and far between, with only the best examples and those with very aggressive and premium ambient cooling solutions capable of this. [...] The Intel Core i9-13900KS is available to buy now at most retailers, with an MSRP of $699. This is $40 cheaper than the previous Core i9-12900KS ($739) that launched last year. Based on current MSRP pricing, the Core i9-13900KS is $110 more than the current Core i9-13900K.

Apple is planning to start using its own custom displays in mobile devices as early as 2024, an effort to reduce its reliance on technology partners like Samsung and LG and bring more components in-house. From a report: The company aims to begin by swapping out the display in the highest-end Apple Watches by the end of next year, according to people with knowledge of the matter. The screens upgrade the current OLED -- organic light-emitting diode -- standard to a technology called microLED, and Apple plans to eventually bring the displays to other devices, including the iPhone. The changes are part of a sweeping effort to replace Apple supplies with homegrown parts, an undertaking that will give the company more control over the design and capabilities of its products. The tech giant has dropped Intel chips in its Mac computers in favor of in-house designs and plans to do the same with the key wireless components in its iPhones.

At CES this week, AMD announced a suite of new chips for notebooks and desktop computers, with one notable announcement being the company's new AMD Ryzen 7040 series of processors for ultrathin notebooks that will compete with Apple's M1 Pro and M2 chips. MacRumors reports: The AMD Ryzen 7040 series of chips are "ultrathin" processors based on the 4nm process, and the highest-end chip part of the family is the Ryzen 9 7940HS. The Ryzen 9 7940HS has eight cores, 16 threads, and 5.2GHz boost speeds. Announcing the new chip, AMD CEO Lisa Su made bold claims about its performance, saying it's up to 30% faster than Apple's M1 Pro chip. In specific tasks, AMD claims the chip is 34% faster in multiprocessing workloads than the M1 Pro and 20% faster than the M2 in AI tasks.

One cornerstone of Apple silicon is energy efficiency, and in that area, AMD claims the new AMD Ryzen 7040 series will offer 30+ hours of video playback in ultrathin notebooks. Built directly into the series of chips is Ryzen AI, a dedicated AI engine embedded in the processor. AMD chips configured with Ryzen AI are 20% faster in AI tasks than Apple's M2 chip while being 50% more energy efficient, according to the company.

To showcase the new chip's performance, AMD compared the performance of a high-end Intel chip, the M1 Pro, and its new Ryzen 9 7940HS processor rendering an object in the popular application Blender. In the time-lapsed video shown on stage, the M1 Pro lags behind the Ryzen 9 7940HS in rendering the object. AMD says it made its performance claims against a MacBook Pro with M1 Pro, 32GB of unified memory, and 1TB of SSD storage running macOS Monterey. The M1 Pro is not Apple's highest-end and most powerful chip for laptops, which is the M1 Max, and AMD did not compare its chip to the M1 Max. After roasting the M1 Pro, Ian Zelbo from FrontPageTech noticed AMD running their CES keynote on multiple 14-inch MacBook Pros. "Obviously these are contracted employees, and it means nothing," he tweeted. "I just always find stuff like this hilarious."

We do too... It's akin to the "Twitter for iPhone" line on tweets that have gotten Android promoters in hot water multiple times over the past several years.

Alienware is unveiling a refreshed lineup of its M- and X-series gaming laptops at CES 2023. Like some other laptop companies, including Razer and Acer, Alienware is shifting focus away from 15- and 17-inch laptops toward thin, powerful 16- and 18-inch models. From a report: The brand is going big with the new M18, an 18-inch model that's being pitched as a desktop replacement. This is actually a resurrection following the M18's previous spec update way back in 2015. The 2023 model will feature Intel's 13th Generation HX CPUs and Nvidia's RTX 4090 mobile graphics card. The latest processors and graphics options from AMD will be available in the M18 later in 2023. Not only is the M18 massive and powerful -- it's a big deal in other ways. It can be configured with an 18-inch QHD Plus screen in the taller 16:10 aspect ratio and set up to include a ton of ports, including two Thunderbolt 4 ports, HDMI 2.1, Ethernet, an SD card reader, and many others. It supports user-upgradeable dual DDR5 RAM slots, and you can also cram up to 9TB of NVMe M.2 storage in it. This model starts at $2,099, but the first configuration it's releasing will cost $2,899.

Intel has finally pulled the proverbial trigger on its non-K series SKUs, with sixteen new Raptor Lake-S series processors for desktops. AnandTech: Varied across a mixture of bare multiplier locked SKUs such as the Core i9-13900 and Core i7-13700 with a TDP of 65 W, Intel has also announced its T series models with a TDP of just 35 W for lower powered computing, including the Core i9-13900T. Furthermore, Intel has launched its Core i3 series family, offering decent performance levels, albeit with just performance (P) cores and no efficiency (E) cores, at a more affordable price starting from $109. Although the overclockable parts typically get consumers' attention when they launch, most of Intel's sales come through its regular non-K parts. Despite not being world record holders regarding performance or overclocking ability, the non-K series SKUs account for most system builders and OEM systems across the entry-level and mid-range offerings.

Intel's non-K launch offerings as part of its Raptor Lake-S architecture all come with a TDP of 65 W or lower, with variants representing the Core i9, Core i7, and Core i5; Intel has also now pulled the trigger on its 13th Gen Core i3 series. Intel has sixteen new desktop processors with varying performance, specification, and price levels, ranging from 24-core (8P+16E) to quad-core (4P+0E) options. Memory support on the Core i9 and Core i7 series includes both DDR5-5600 and DDR4-3200, while the new Core i5 and Core i3 series support DDR5-4800 and DDR4-3200 as per JEDEC specifications. There are three new Intel 13th Gen Core i9 series processors to select from, starting at $549 with the Core i9-13900. All Core i9 series non-K parts include 8P+16E cores for 32 threads, and 36 MB of Intel Smart L3 cache, with the Core i9-13900 ($549) and Core i9-13900F ($524) sharing the same 5.6 GHz turbo clock speed and a base frequency of 3.3 GHz on the performance (P) cores. Both models also include a base TDP of 65 W and a turbo TDP of 219 W, which is plenty of power budget for turbo clock speeds on both the P and E cores. The only caveat is that the Core i9-13900F doesn't include Intel's UHD 770 integrated graphics (32 EUs); consequently, it has a $25 lower MSRP.

The third of Intel's Core i9 non-K series chips is the Core i9-13900T, with the T signifying that it's a 35 W part. A lower power envelope means it sacrifices plenty of MHz to account for the drop in power. The Core i9-13900 has a P-core base frequency of 1.1 GHz, with a turbo clock speed of up to 5.3 GHz; the E-core specifications are similar, with a base frequency of 800 MHz and a turbo of 3.9 GHz. Even though the Core i9-13900T ($549) comes with a 35 W base TDP, it has a turbo TDP of 106 W. Moving onto the Core i5 family, Intel has three new Raptor Lake-S desktop processors, including two 65 W and one T series (35 W) part. All three include 30 MB of Intel The Core i7-13700 and Core i7-13700F both feature a P-core turbo clock speed of 5.2 GHz, while the restrictions in power mean that the P-core base frequency sits at just 2.1 GHz. For the efficiency (E) cores, this means that they have a base frequency of 1.5 GHz and a turbo clock speed of 4.1 GHz, while both conform to Intel's interpretation of 65 W; they both have a turbo TDP of 219 W. The Core i7-13700T, as per the specifications, has a base TDP of 35 W, but it has a turbo TDP of 106 W. As with other T-series family products, the lower TDP puts constraints on raw frequency, with a P-core base frequency of just 1.4 GHz, but the eight performance cores boost to 4.9 GHz, while the eight efficiency cores turbo up to 3.6 GHz. It shares the same level of 30 MB of L3 cache as the other Raptor Lake-S desktop Core i7 processors and includes Intel's UHD 770 integrated graphics chip.

There's been "an enormous ramp-up in U.S. chip-making plans" over the last 18 months, reports the New York Times. For example:

- In September Intel pledged $20 billion for two chip factories in Ohio
- Micron expects to spend at least that amount on a new manufacturing site in Syracuse, New York.
- Taiwan Semiconductor Manufacturing Company plans to invest $40 billion in Phoenix.

"The boom has implications for global technological leadership and geopolitics, with the United States aiming to prevent China from becoming an advanced power in chips..." Across the U.S., more than 35 companies have pledged nearly $200 billion for manufacturing projects related to chips since the spring of 2020, according to the Semiconductor Industry Association, a trade group. The money is set to be spent in 16 states, including Texas, Arizona and New York on 23 new chip factories, the expansion of nine plants, and investments from companies supplying equipment and materials to the industry. The push is one facet of an industrial policy initiative by the Biden administration, which is dangling at least $76 billion in grants, tax credits and other subsidies to encourage domestic chip production....

The new U.S. production efforts may correct some of these imbalances, industry executives said — but only up to a point. The new chip factories would take years to build and might not be able to offer the industry's most advanced manufacturing technology when they begin operations. Companies could also delay or cancel the projects if they aren't awarded sufficient subsidies by the White House. And a severe shortage in skills may undercut the boom, as the complex factories need many more engineers than the number of students who are graduating from U.S. colleges and universities....

A $50 billion government investment is likely to prompt corporate spending that would take the U.S. share of global production to as much as 14 percent by 2030, according to a Boston Consulting Group study in 2020 that was commissioned by the Semiconductor Industry Association. "It really does put us in the game for the first time in decades," said John Neuffer, the association's president, who added that the estimate may be conservative because Congress approved $76 billion in subsidies in a piece of legislation known as the CHIPS Act.
The article also cites predictions of 40,000 new jobs (made by the Semiconductor Industry Association) in exploring the possibility of a U.S. "talent shortage."

"Intel, responding to the issue, plans to invest $100 million to spur training and research at universities, community colleges and other technical educators."

Demand for graphics cards significantly increased during the pandemic as some people spent more time at home playing games, whereas others tried to mine Ethereum to get some cash. But it looks like now that the world has re-opened and Ethereum mining on GPUs is dead, demand for desktop discrete GPUs has dropped dramatically. From a report: In fact, shipments of discrete graphics cards hit a ~20-year low in Q3 2022, according to data from Jon Peddie Research. The industry shipped around 6.9 million standalone graphics boards for desktop PCs -- including the best graphics cards for gaming -- and a similar number of discrete GPUs for notebooks in the third quarter.

In total, AMD, Intel, and Nvidia shipped around 14 million standalone graphics processors for desktops and laptops, down 42% year-over-year based on data from JPR. Meanwhile, shipments of integrated GPUs totaled around 61.5 million units in Q3 2022. In fact, 6.9 million desktop discrete add-in-boards (AIBs) is the lowest number of graphics cards shipped since at least Q3 2005 and, keeping in mind sales of standalone AIBs were strong in the early 2000s as integrated GPUs were not good enough back then, it is safe to say that in Q3 2022 shipments of desktop graphics boards hit at least a 20-year low.

HandBrake, the popular free and open source video transcoder, has been updated to version 1.6.0. This major point upgrade is notable for facilitating AV1 video encoding for the first time in a general release. Moreover, those with Intel Quick Sync Video (QSV) enabled processors, and those with Intel Arc GPUs will be able to encode AV1 video with hardware acceleration. From a report: HandBrake 1.6.0 can encode AV1 videos on any of its supported systems. In the current release its SVT-AV1 encoder offers the widest support, encoding on your processor through software. However, those with Intel QSV supporting CPUs or discrete Arc graphics can use the QSV-AV1 encoder for hardware accelerated processing. QSV isn't supported if your CPU is an 'F' suffixed model (i.e. it doesn't have an iGPU), or it is older than the Skylake generation. If you are lucky enough to have multiple QSV accelerators in your system, support for Intel Deep Link Hyper Encode should accelerate processing further. While AMD and Nvidia have AV1 encoders available for their latest GPUs, they currently aren't integrated with HandBrake. AV1 video is set to become the dominant codec across app-based streaming services and the wider internet, offering attractions such as; an open and royalty-free architecture, improved compression enabling efficient 8K video streaming, and support for the newest HDR standards.

Intel and SoftBank-backed VLSI Technology have agreed to end a $4 billion patent dispute, according to documents filed in Delaware District Court this week. From a report: The decision marks a victory for Intel, which has already lost $3 billion in failed patent disputes to VLSI over the past few years. The case in question dates back to 2018 and alleged that Intel had infringed on five VLSI-owned patents governing things like secure communications, power optimization and delivery, and flip-chip interconnects.

If VLSI sounds familiar, that's because the company has been lurking around the semiconductor industry in one shape or form since the late '70s. The company originally made ASICs before it was acquired by Philips Electronics and later spun off under NXP. But despite any early successes in chipmaking, VLSI is now owned by SoftBank's Fortress Investment Group, and appears to exist solely to sue chipmakers it believes have violated its intellectual property -- in other words, it's a patent troll. The decision to call it quits comes after nearly five years of litigation. Tuesday, Intel and VLSI released a joint filing in which Intel and VLSI mutually agreed to dismiss the case and resolve all disputes over Intel's use of the aforementioned patents. Critically, VLSI has done so with prejudice. As we understand it, this means the company can't refile the case.

The number of GPU startups in China is extraordinary as the country tries to gain AI prowess as well as semiconductor sovereignty, according to a new report from Jon Peddie Research. From a report: In addition, the number of GPU makers grew worldwide in recent years as demand for artificial intelligence (AI), high-performance computing (HPC), and graphics processing increased at a rather unprecedented rate. When it comes to discrete graphics for PCs, AMD and Nvidia maintain lead, whereas Intel is trying to catch up.

Tens of companies developed graphics cards and discrete graphics processors in the 1980s and the 1990s, but cut-throat competition for the highest performance in 3D games drove the vast majority of them out of business. By 2010, only AMD and Nvidia could offer competitive standalone GPUs for gaming and compute, whereas others focused either on integrated GPUs or GPU IP. The mid-2010s found the number of China-based PC GPU developers increasing rapidly, fueled by the country's push for tech self-sufficiency as well as the advent of AI and HPC as high-tech megatrends.

In total, there are 18 companies developing and producing GPUs, according to Jon Peddie Research. There are two companies that develop SoC-bound GPUs primarily with smartphones and notebooks in mind, there are six GPU IP providers, and there are 11 GPU developers focused on GPUs for PCs and datacenters, including AMD, Intel, and Nvidia, which design graphics cards that end up in our list of the best graphics cards. In fact, if we added other China-based companies like Biren Technology and Tianshu Zhixin to the list, there would be even more GPU designers. However, Biren and Tianshu Zhixin are solely focused on AI and HPC for now, so JPR does not consider them GPU developers.

An anonymous reader shares a report: I reviewed a number of laptops in 2022 across consumer, workstation, gaming, business, Chromebook, and everywhere else. I touched all of the major brands. But I had a particular focus on ultraportables this year -- that is, thin and light devices that people buy to use, say, on their couch at home -- because, with Apple's MacBooks in such a dominant position, many eyes have been on their competitors on the Windows side. For many of these models, I found myself writing the same review over and over and over. They were generally good. They performed well. But their battery life was bad.

What these laptops had in common is that they were all powered by the Intel P-series. Without getting too into the weeds here, Intel processors have, in the past, included H-series processors -- powerful chips that you'll find in gaming laptops and workstations -- and U-series processors for thinner, lighter devices. (There was also a G-series, which was this whole other thing, for a couple of years.) But the Intel 12th Generation of mobile chips (that is, the batch of chips that Intel released this year) has a new letter: the P-series. The P-series is supposed to sit between the power-hungry H-series and the power-efficient U-series; the hope was that it would combine H-series power with U-series efficiency.

And then many -- a great many -- of this year's top ultraportable laptops got the P-series: big-screeners like the LG Gram 17; modular devices like the Framework Laptop; business notebooks like the ThinkPad X1 Yoga Gen 7; premium ultraportables like the Acer Swift 5, the Lenovo Yoga 9i, the Samsung Galaxy Book2 Pro, and the Dell XPS 13 Plus. The problem was that, in reality, the P-series was just a slightly less powerful H-series chip, which Intel had slapped an "ultraportable" label onto. It was identical to the H-series in core count and architecture, but it was supposed to draw slightly less power.

Intel announced today that it would split its AXG graphics group to separately address the gaming and data center markets by placing it under two other business units. Raja Koduri, currently the Executive Vice President of the AXG business unit, will return to his previous role as an Intel Chief Architect. From a report: "Discrete graphics and accelerated computing are critical growth engines for Intel. With our flagship products now in production, we are evolving our structure to accelerate and scale their impact and drive go-to-market strategies with a unified voice to customers. This includes our consumer graphics teams joining our client computing group, and our accelerated computing teams joining our datacenter and AI group. In addition, Raja Koduri will return to the Intel Chief Architect role to focus on our growing efforts across CPU, GPU and AI, and accelerating high priority technical programs," Intel said.

We spoke with Intel, and the company assures us that it remains fully committed to its existing roadmap of Arc consumer discrete GPUs, meaning it intends to launch the second-gen Battlemage and third-gen Celestial gaming GPUs as planned. Those GPUs will join the recently launched Alchemist series, which will also continue to be supported.

An anonymous reader quotes a report from Bloomberg, written by Mark Gurman: The new high-end Mac Pro with Apple silicon is behind schedule, and you can blame changes to the company's chip and manufacturing plans. When Apple announced plans in June 2020 to transition away from Intel processors to Mac chips designed in-house, the company said the move would take about two years. Now at the tail end of 2022, it's clear that Apple has missed its self-imposed deadline for completing the shift. In addition to not offering a Mac Pro with Apple silicon, the company still only sells the high-end version of the Mac mini desktop in an Intel flavor. While Apple has said little to nothing about its future Mac desktops or the reasons behind the holdup, the company continues to actively test an all-new Mac Pro and an M2 Pro-based Mac mini to replace the remaining Intel models. Apple had aimed to introduce the new Mac Pro by now, but the high-end machine has been held up for a number of reasons, including multiple changes to its features, a significant shift in the company's plans for high-end processors and a potential relocation of its manufacturing.

When Apple first set out to build a replacement for the Intel Mac Pro, it planned a machine with a processor based on the original M1 chip. The approach called for two main configurations: one chip equal to the power of two M1 Max processors -- the highest-end MacBook Pro chip -- and another equal to four M1 Max components combined. The dual M1 Max chip ended up first launching in the Mac Studio as the M1 Ultra, and Apple decided to push back the Mac Pro to the M2 generation. The company then planned for the Mac Pro to come in two configurations: an M2 Ultra version and a double-M2 Ultra that I've dubbed the "M2 Extreme." The M2 Ultra chip is destined to have some serious specifications for professional users, including up to 24 CPU cores, 76 graphics cores and the ability to top out the machine with at least 192 gigabytes of memory. An M2 Extreme chip would have doubled that to 48 CPU cores and 152 graphics cores. But here's the bad news: The company has likely scrapped that higher-end configuration, which may disappoint Apple's most demanding users -- the photographers, editors and programmers who prize that kind of computing power.

The company made the decision because of both the complexity and cost of producing a processor that is essentially four M2 Max chips fused together. It also will help Apple and partner Taiwan Semiconductor Manufacturing Co. save chip-production resources for higher-volume machines. Moreover, there are concerns about how much consumers are willing to spend. Using the highest-end M1 Ultra chip pushes the Mac Studio up to $5,000 -- only $1,000 less than the current Mac Pro. That's $3,000 more than the M1 Max Mac Studio. Based on Apple's current pricing structure, an M2 Extreme version of a Mac Pro would probably cost at least $10,000 -- without any other upgrades -- making it an extraordinarily niche product that likely isn't worth the development costs, engineering resources and production bandwidth it would require. Instead, the Mac Pro is expected to rely on a new-generation M2 Ultra chip (rather than the M1 Ultra) and will retain one of its hallmark features: easy expandability for additional memory, storage and other components. Gurman says the Mac Mini update "will come in regular M2 and M2 Pro variations, while new 14-inch and 16-inch MacBook Pros are arriving early next year with M2 Pro and M2 Max options." A high-end iMac Pro with Apple silicon is also in the works, "but that machine has suffered internal delays for similar reasons as the Mac Pro," he notes.

In addition, Gurman says Apple is "working on multiple new external monitors [...], including an update to the Pro Display XDR that was launching alongside the Intel Mac Pro in 2019." The new monitors will also include Apple silicon.

An anonymous reader quotes a report from Ars Technica: Dell continues tinkering with what it hopes to be a repairable laptop like the Framework Laptop. Last year, it showed off Concept Luna, a clamshell designed to easily disassemble for easy repairs, upgrades, and harvested components. This year, Dell showed the press an updated Concept Luna that could support more power while being even simpler to dismantle. The vendor is also exploring how to automate the process, from disassembly to parts diagnostics, on a broad scale. Dell's Concept Luna laptop is comparable in size to a Latitude with some Dell XPS 13 Plus-like stylings. In person, it looked similar to the Concept Luna demoed last year, including appearing to be a functioning PC. But Dell's representative was able to open this year's version up and pull out internal parts much more rapidly -- well under 60 seconds.

The computer was easier to take apart because it doesn't have screws (last year's Concept Luna had four). Dell's rep simply stuck a pin (it could be anything that fits, they said) into a hole in the security lock slot on the right side of the system's deck. That allowed Dell's rep to pull off the keystone north of the keyboard and then slide the keyboard up and out. Once the system was open, the speakers, fan, motherboard, and battery were removed instantly thanks to pop-out modules, which, Dell said, are recyclable. The concept laptop also got rid of the cable connecting the battery, so there are no cables, adhesives, or other types of connectors.

Dell built Concept Luna with simple display upgrades and repairs in mind as well. Dell's spokesperson promptly removed the system's LCD by inserting a pin into a hole in a keystone south of the screen and then took off the keystone, releasing a latch underneath the piece, and plucked the display off the chassis. There's also potential for better accommodations for beefier components in the updated Concept Luna. Last year's version used passive cooling, while the new one has a fan. The fans lock into the motherboard, keeping it in place. A tech giant like Dell releasing something like Concept Luna could certainly give the younger Framework a run for its money, but Dell still isn't talking about releasing a laptop with Concept Luna's repairability. And it could ultimately decide not to. However, in addition to advancing the laptop's design this year, Dell also looked into automation techniques that could further this concept on a scale that could extend beyond a single product.

Anyone remember the high-end commercial UNIX workstations from a few decades ago — like from companies like IBM, DEC, SGI, and Sun Microsystems?

Today OSnews looked back — but also explored what happens when you try to buy one today> : As x86 became ever more powerful and versatile, and with the rise of Linux as a capable UNIX replacement and the adoption of the NT-based versions of Windows, the days of the UNIX workstations were numbered. A few years into the new millennium, virtually all traditional UNIX vendors had ended production of their workstations and in some cases even their associated architectures, with a lacklustre collective effort to move over to Intel's Itanium — which didn't exactly go anywhere and is now nothing more than a sour footnote in computing history.

Approaching roughly 2010, all the UNIX workstations had disappeared.... and by now, they're all pretty much dead (save for Solaris). Users and industries moved on to x86 on the hardware side, and Linux, Windows, and in some cases, Mac OS X on the software side.... Over the past few years, I have come to learn that If you want to get into buying, using, and learning from UNIX workstations today, you'll run into various problems which can roughly be filed into three main categories: hardware availability, operating system availability, and third party software availability.
Their article details their own attempts to buy one over the years, ultimately concluding the experience "left me bitter and frustrated that so much knowledge — in the form of documentation, software, tutorials, drivers, and so on — is disappearing before our very eyes." Shortsightedness and disinterest in their own heritage by corporations, big and small, is destroying entire swaths of software, and as more years pass by, it will get ever harder to get any of these things back up and running.... As for all the third-party software — well, I'm afraid it's too late for that already. Chasing down the rightsholders is already an incredibly difficult task, and even if you do find them, they are probably not interested in helping you, and even if by some miracle they are, they most likely no longer even have the ability to generate the required licenses or release versions with the licensing ripped out. Stuff like Pro/ENGINEER and SoftWindows for UNIX are most likely gone forever....

Software is dying off at an alarming rate, and I fear there's no turning the tide of this mass extinction.
The article also wonders why companies like HPE don't just "dump some ISO files" onto an FTP server, along with patch depots and documentation. "This stuff has no commercial value, they're not losing any sales, and it will barely affect their bottom line.

The next wave of Moore's Law will rely on a developing concept called system technology co-optimization, Ann B. Kelleher, general manager of technology development at Intel told IEEE Spectrum in an interview ahead of her plenary talk at the 2022 IEEE Electron Device Meeting. From a report: "Moore's Law is about increasing the integration of functions," says Kelleher. "As we look forward into the next 10 to 20 years, there's a pipeline full of innovation" that will continue the cadence of improved products every two years. That path includes the usual continued improvements in semiconductor processes and design, but system technology co-optimization (STCO) will make the biggest difference. Kelleher calls it an "outside-in" manner of development. It starts with the workload a product needs to support and its software, then works down to system architecture, then what type of silicon must be within a package, and finally down to the semiconductor manufacturing process. "With system technology co-optimization, it means all the pieces are optimized together so that you're getting your best answer for the end product," she says.

STCO is an option now in large part because advanced packaging, such as 3D integration, is allowing the high-bandwidth connection of chiplets -- small, functional chips -- inside a single package. This means that what would once be functions on a single chip can be disaggregated onto dedicated chiplets, which can each then be made using the most optimal semiconductor process technology. For example, Kelleher points out in her plenary that high-performance computing demands a large amount of cache memory per processor core, but chipmaker's ability to shrink SRAM is not proceeding at the same pace as the scaling down of logic. So it makes sense to build SRAM caches and compute cores as separate chiplets using different process technology and then stitch them together using 3D integration. A key example of STCO in action, says Kelleher, is the Ponte Vecchio processor at the heart of the Aurora supercomputer. It's composed of 47 active chiplets (as well as 8 blanks for thermal conduction). These are stitched together using both advanced horizontal connections (2.5 packaging tech) and 3D stacking. "It brings together silicon from different fabs and enables them to come together so that the system is able to perform against the workload that it's designed for," she says.