Asus on Thursday officially introduced the ROG Ally, its first handheld gaming PC. With numerous handheld gaming systems around, most notably Steam Deck, Asus needed something special to be successful and fulfill the promise of the ROG brand. To that end, the ROG Ally promises a unique combination of performance enabled by AMD's latest mobile CPU, high compatibility due to usage of Windows 11, portability, and other features.

Performance: To Extreme, or Not to Extreme?

First teased by Asus last month, the ROG Ally is the company's effort to break into the handheld gaming PC space, which Valve has essentially broken open in the past year with the Steam Deck.

When developing ROG Ally, Asus wanted to build a no-compromise machine that would bring the performance of mobile PCs the portability that comes with handheld device. This is where AMD's recently-launched Zen 4-based Ryzen Z1 and Ryzen Z1 Extreme SoCs, which are aimed specifically at ultra-portable devices, come into play.

Based on AMD's 4nm Phoenix silicon, the eight-core Ryzen Z1 Extreme processor and its 12 CU RDNA 3-based GPU resembles the company's Ryzen 7 7840U CPU. Meanwhile Asus is also offering a version of Ally using the lower-tier Z1 chip, which still uses eight CPU cores and pairs that with a 4 CU GPU. On paper, the Z1 Extreme chip is significantly more powerful in graphics tasks as a result (~3x), however in practice the chips are closer, as thermal and memory bandwidth limits keep the Extreme chip from running too far ahead.

Speaking of graphics performance, it should be noted that Asus's ROG Ally console is equipped with the ROG XG Mobile connector (a PCIe 3.0 x8 for data and a USB-C for power and USB connections) that can be used to connect an Asus ROG XG Mobile eGFX dock with the handheld. The XG docks come with a range of GPUs installed, up to a GeForce RTX 4090 Laptop GPU. The XG dock essentially transforms ROG Ally into a high-performance gaming system, albeit by supplanting much of its on-board functionality. The fact that Asus offers eGFX capability right out-of-box is a significant feature differentiator for the ROG Ally, though be prepared to invest the $1999.99 if you want the top-end GeForce RTX 4090 Laptop-equipped XG dock.

Both versions of ROG Ally will come with 16GB of LPDDR5-6400 memory and a 512GB SSD in an M.2-2230 form-factor with a PCIe 4.0 interface. While replacing the M.2 drive is reportedly a relatively easy task, for those who want to expand storage space without opening anything up, the console also has an UHS-II-compliant microSD card slot.

Display: Full-HD at 120 Hz

The ROG Ally is not only the first handheld with the Ryzen Z1 Extreme CPU, but will also be among the first portable game consoles with a 1920x1080 resolution 7-inch display; and one that supports a maximum refresh rate of 120 Hz, no less. The Gorilla Glass Victus-covered display uses an IPS-class panel with a peak luminance of 500 nits as well as Dolby Vision HDR support to make games more appealing.

In addition to Dolby Vision HDR-badged display, the Asus ROG Ally also has Dolby Atmos-certified audio subsystem with Smart Amp speakers and noise cancelation technology. 

Ergonomics: 600 Grams and All the Controls

When it comes to mobile devices, ergonomics is crucial. Yet, it is pretty hard to design a handheld game console that essentially uses laptop-class silicon with all of its peculiarities. When Asus began work on its ROG Ally, it asked mobile gamers about what they think was the most important feature for their portable console and apparently it was weight. So Asus set about deigning a device that would weigh around 600 grams and would be comfortable to use.

"When we go through survey with our focus group, the number one thing that they wanted was a balanced weight handheld device," said Shawn Yen, vice president of Asus's Gaming Business Unit responsible for ROG products. "The target was 600 grams because the current handheld devices in the market today are too heavy. It is not something that they can engage for a very long period of time. So, their game time got cut down because it is not comfortable. So, uh, when we first thought about the design target for ROG Ally, we were thinking about a device that can get into gamers' hands for hours of fun time."

The display and chassis are among the heaviest components of virtually all mobile devices, so there is little that can be done about those. But in a bid to optimize the weight and distribute it across the device, the company had to implement a very well thought motherboard design, and use anti-gravity heat pipes to ensure proper cooling at all times without using too many of them as this increases weight. Meanwhile, Asus still had to use two fans and a radiator with 0.1 mm ultra-thin fins to ensure that the CPU is cooled down properly as it still can dissipate up to 30W of heat. To further optimize weight, Asus opted for a polycarbonate chassis.

Since Asus ROG Ally is essentially a Windows 11-based PC albeit in a portable game console form factor, the company had to incorporate all the pads and buttons featured on conventional gamepads and some more controls for Windows (e.g., touchscreen) and ROG Ally-specific things like Armor Crate game launcher and two macro buttons. It's also worth noting that, seemingly because of the use of Windows 11, the Ally is not capable of consistently suspending games while it sleeps, a notable difference compared to other handheld consoles.

Meanwhile, the trade-off to hitting their weight target while still using a relatively powerful SoC has been battery life. The Ally comes with a 40Wh batter, and Asus officially advertises the handheld as offering up to 2 hours of battery life in heavy gaming workloads. Early reviews, in turn, have matched this, if not coming in below 2 hours in some cases. The higher-resolution display and high-performance AMD CPU are both key differentiating factors of the Ally, but these parts come at a high power cost.

Vast Connectivity

Being a PC, the ROG Ally is poised to offer connectivity that one comes to expect from a portable computer. Therefore, the unit features a Wi-Fi 6E and Bluetooth adapter for connectivity, it includes a MicroSD card slot for additional storage, a USB Type-C port for both charging and display output, an ROG XG Mobile connector for external GPUs, and a TRRS audio connector for headsets.

The Price

The ROG Ally with AMD's Ryzen Z1 Extreme CPU is set to be launched globally on June 13, 2023, at a price point of $699.99. Meanwhile the non-extreme Z1 version of the Ally has been lited for $599.99, though no release date has been set. The first reviews are already out, so Asus is giving potential customers a long lead time to evaluate the console before it's released next month.

Asus has expanded the company's GeForce RTX 40-series product portfolio with two new RTX 4090 graphics cards. The ROG Strix LC GeForce RTX 4090 and TUF Gaming GeForce RTX 4090 OC, which are available in regular and OC editions, have arrived to compete in the high-end segment. What makes these cards notable, in turn, is their reduced size: the new cards are physically smaller than Asus' early RTX 4090 offerings, as well as many of the competitors on the market.

The GeForce RTX 4090 is a 450W gaming graphics card, with large coolers to match. Even NVIDIA's hard-to-get GeForce RTX 4090 Founders Edition is a triple-slot graphics card, and air-cooled AIB cards tend to be larger still. So for the size-conscious gamer, this leaves liquid cooled cards, which brings us to Asus's new ROG Strix LC GeForce RTX 4090. The closed-loop card moves a lot of its bulk off to an attached 240 mm radiator block, bringing the card itself down to 2.6-slots wide.

The ROG Strix LC GeForce RTX 4090's hybrid cooling system packs a cold plate that cools the large AD102 GPU and neighboring GDDR6X memory chips. The heat is transferred to the 240 mm radiator through 560 mm tubing, so there won't be an issue with large cases. A low-profile heatsink with a blower-style cooling fan keeps the other power delivery components cool. Meanwhile the radiator itself is equipped with a pair of 120 mm ARGB cooling fans are present to dissipate the heat once it gets there.

Asus's other new RTX 4090 card, the air-cooled TUF Gaming GeForce RTX 4090 OG, is a unique case of its own. Technically, it's a new SKU; however, the graphics card reuses the TUF Gaming cooler from the TUF Gaming GeForce RTX 3090 Ti.

This is notable because the TUF cooler used on the 3090 Ti was a good bit smaller than Asus's first RTX 4090 cooler. The net result is that these changes bring the new OG card's width from 3.65-slots (and arguably, wide enough that you need to leave a 4th slot open for air flow) down to 3.2 slots - just enough room for proper airflow if the neighboring 4th sot is occupied. Altogether, the OG model is smaller in every dimension, shaving off 6% of its height, 7% of its length, and 13% of its width. Asus doesn't list the weight of its graphics cards, so we cannot comment on whether the new OG version has lost weight.

By most accounts, Asus's current RTX 4090 cooler is highly effective – it's just also really big. So offering a separate SKU with a smaller cooler makes a good deal of sense, especially given how popular NVIDIA's true triple-slot Founders Edition card has been. The smaller TUF cooler is rated for the same 450W TDP as the larger TUF 4090 cooler, but, as always, there may be performance/acoustic tradeoffs involved.

There's one other change that Asus doesn't advertise with the TUF Gaming GeForce RTX 4090 OG. The renders on the product page show the graphics card with a longer PCB. One of the advantages of the more compact PCB on the previous model was that it permitted Asus (and NVIDIA) to vent heat out of the back side of the card, as well as to optimize the trace layouts and component placement. Meanwhile, with the longer PCB, Asus relocated the 16-pin power connector. Instead of being placed in the middle, the power connector is on the farther right side.

Gallery: Asus GeForce RTX 4090 GPUs

Between the two new cards, the ROG Strix LC GeForce RTX 4090 ends up with the edge in clockspeeds, flaunting boost clock speeds up to 2,640 MHz when in its highest performance mode. Meanwhile, the TUF Gaming GeForce RTX 4090 OG series have the same clock speeds as the vanilla models, with a rated boost clock of 2520 MHz stock and 2595 MHz when the OC card is in its highest mode. In addition, the ROG Strix LC GeForce RTX 4090 and TUF Gaming GeForce RTX 4090 OG have other attributes in common, including using a single 16-pin power connector and a display output layout consisting of two HDMI 2.1a ports and three DisplayPort 1.4a outputs.

Asus hasn't revealed the pricing or availability of the new graphics cards. For reference, the TUF Gaming GeForce RTX 4090 and OC Edition retail for $1,599 and $1,799, respectively. The OG counterparts likely have similar price tags. Meanwhile, we'd expect the ROG Strix LC GeForce RTX 4090 to carry a more considerable premium due to the AIO liquid cooling design.

Although E Ink technology has remained a largely niche display tech over the past decade, it's none the less excelled in that role. The electrophoretic technology closely approximates paper, providing significant power advantages versus traditional emissive displays, not to mention making it significantly easier on readers' eyes in some cases. And while the limitations of the technology make it unsuitable for use as a primary desktop display, Phillips thinks there's still a market for it as a secondary display. To that end, Philips this week has introduced their novel, business-oriented Dual Screen Display, which combines both an LCD panel and and E Ink panel into a single display, with the aim of capturing the benefits of both technologies.

The Philips Dual Screen Display (24B1D5600/96) is a single display that integrates both a 23.8-inch 2560x1440 IPS panel as well as a 13.3-inch, greyscale 1200x1600 resolution E Ink display. With each display operating independently, the idea is similar to previous concepts of multi-panel monitors; however Phillips is taking things in a different direction by using an E Ink display as a second panel – combining two otherwise very different display technologies into a single product. By offering an E Ink panel in this product, Phillips is looking to court the market for users who would prefer the reduced eye strain of an E Ink display, but are working at a desktop computer, where an E Ink display would not be viable as a primary monitor.

As you might expect from the basic layout of the monitor, the primary panel is a rather typical office display that's designed for video and productivity applications – essentially anything where you need a modern, full color LCD. The secondary E Ink display, on the other hand, is a greyscale screen whose strength is the lack of flicker that comes from not being backlit by a PWM light. Both screens act independently, but since they are encased into the same chassis, they are meant to work together. For example, the secondary monitor can display supplementary information in text form, whereas the primary monitor can display photos.

Ultimately, Philips is pitching the display on the idea that the secondary screen can reduce the eye strain of the viewer while viewing documents. It's a simple enough concept, but one that requires buyers to overlook the trade-offs of E Ink, and the potential drawbacks of having two dissimilar displays directly next to each other.

Under the hood, the LCD panel on the Deal Screen Display is an unremarkable office-grade display. Phillips is using 23.8-inch anti-glare 6-bit + Hi FRC IPS panel with a 2560x1440 resolution, which can hit a maximum brightness of 250 nits while delivering 178-degree viewing angles. Meanwhile, the E Ink panel is a 13.3-inch 4-bit greyscale electrophoretic panel, with a resolution of 1200x1600. Notably here, there is no backlighting; the E Ink panel is meant to be environmentally lit (e.g. office lighting) to truly minimize eye strain.

When it comes to connectivity, the primary screen is equipped with a DisplayPort 1.2 and a USB Type-C input (with DP Alt mode and USB Power Delivery support), a USB hub, and a GbE adapter. Meanwhile, the secondary screen connects to host using a USB Type-C connector that also supports DP Alt Mode, and Power Delivery.

The Philips Dual Screen Display has a rather sleek stand which can adjust height, tilt, and swivel. It makes the whole unit look like one monitor rather than like two separate screens. Though to be sure, the E Ink portion of the display can be angled independently from the LCD panel, allowing the fairly wide monitor to contour to a user's field of view a bit better.

When it comes to pricing, Philips's Dual Screen Display is available in China for $850 (according to Liliputing), which looks quite expensive for a 24-inch IPS LCD and a 13.3-inch secondary screen. Though as this is a rather unique product, it is not surprising that it is sold at a premium.

Samsung Foundry is set to detail its second generation 3 nm-class fabrication technology as well as its performance-enhanced 4 nm-class manufacturing process at the upcoming upcoming 2023 Symposium on VLSI Technology and Circuits in Kyoto, Japan. Both technologies are important for the contract maker of chips as SF3 (3GAP) promises to offer tangible improvements for mobile and SoCs, whereas SF4X (N4HPC) is designed specifically for the most demanding high-performance computing (HPC) applications.

2^nd Generation 3 nm Node with GAA Transistors

Samsung's upcoming SF3 (3GAP) process technology is an enhanced version of the company's SF3E (3GAE) fabrication process, and relies on its second-generation gate-all-around transistors – which the company calls Multi-Bridge-Channel field-effect transistors (MBCFETs). The node promises additional process optimizations, though the foundry prefers not to compare SF3 with SF3E. Compared to its direct predecessor, SF4 (4LPP, 4nm-class, low power plus), SF3 claims a 22% performance boost at the same power and complexity or a 34% power reduction at the same clocks and transistor count, as well as a 21% logic area reduction. Though it is unclear whether the company has achieved any scaling for SRAM and analogue circuits.

In addition, Samsung claims that SF3 will provide additional design flexibility facilitated by varying nanosheet (NS) channel widths of the MBCFET device within the same cell type. Curiously, variable channel width is a feature of GAA transistors that has been discussed for years, so the way Samsung is phrasing it in context of SF3 might mean that SF3E does not support it.

Samsung's Ealiest 4nm Node: SF4E (IEDM 2021)

Thus far neither Samsung LSI, the conglomerate's chip development arm, nor other customers of Samsung Foundry have formally introduced a single highly-complex processor mass produced on SF3E/3GAE process technology. In fact, it looks like the only publicly-acknowledged application that uses the industry's first 3 nm-class fabrication process is a cryptocurrency mining chip, according to TrendForce. This is not particularly surprising as usage of Samsung's 'early' nodes is typically quite limited. 

By contrast, Samsung's 'plus' technologies are typically used by a wide range of customers, so the company's SF3 (3GAP) process is likely to see much higher volumes when it becomes available sometime in 2024.

SF4X for Ultra-High-Performance Applications

In addition to SF3, which is designed for a variety of possible use cases, Samsung Foundry is prepping its SF4X (4HPC, 4 nm-class high-performance computing) designed for performance-demanding applications like datacenter-oriented CPUs and GPUs.

To address such chips, Samsung's SF4X offers a performance boost of 10% coupled with a 23% power reduction. Samsung doesn't explicitly specify what process node that comparison is being made against, but presumably, this is against their default SF4 (4LPP) fabrication technology. To achieve this, Samsung redesigned transistors' source and drain after reassessing their stresses (presumably under high loads), performed further transistor-level design-technology co-optimization (T-DTCO), and introduced a new middle-of-line (MOL) scheme. 

The new MOL enabled SF4X to offer a silicon-proven CPU minimum voltage (Vmin) of 60mV, a 10% decrease in the variation of off-state current (IDDQ), guaranteed high voltage (Vdd) operation at over 1V without performance degradation, and an improved SRAM process margin.

Samsung's SF4X will be a rival for TSMC's N4P and N4X nodes, which are due in 2024 and 2025 respectively. Based on claim specificaitons alone, it is hard to tell which technology will offer the best combination of performance, power, transistor density, efficiency, and cost. That said, SF4X will be Samsung's first node in the recent years that was specifically architected with HPC in mind, which implies that Samsung has (or is expecting) enough customer demand to make it worth their time.

NVIDIA is launching a new game bundle for its latest generaiton GeForce RTX 40-series graphics cards and OEM systems. This time, NVIDIA has teamed up with Activision Blizzard to offer a free copy of the latest iteration of their wildly popular action RPG series, Diablo IV.

This promotion will run globally, starting now and running through June 16, 2023. For more than a month, customers purchasing GeForce RTX 4090, 4080, 4070 Ti, 4070 graphics cards or desktops containing one of them from various vendors will get a free digital download code of Diablo IV Standard Edition on Battle.net. The code for the title must be redeemed before July 13, 2023.

For NVIDIA, Diablo IV will also be a technology showcase, as it is set to support the DLSS 3 upscaling technology as well as the Reflex latency cutting out-of-box at launch. Ray tracing is also slated to be added at some point after the game launches. At retail pricing, Activision Blizzard's Diablo IV Standard Edition costs $69.99 at Battle.net, though NVIDIA is undoubtedly getting a bulk deal.

It should be noted that this latest game bundle is just for NVIDIA's RTX 40 series desktop cards. Unlike the since-expired Redfall bundle, NVIDIA is not offering Diablo IV (or any other games) with GeForce-based laptops. Nor are any remaining GeForce RTX 30 series producted covered.

Diablo IV will officially release on June 4, 2023.

Source: NVIDIA

In a brief note posted to its investor relations portal this morning, AMD has announced that they will be holding a special AI and data center-centric event on June 13^th. Dubbed the “AMD Data Center and AI Technology Premiere”, the live event is slated to be hosted by CEO Dr. Lisa Su, and will be focusing on AMD’s AI and data center product portfolios – with a particular spotlight on AMD’s expanded product portfolio and plans for growing out these market segments.

The very brief announcement doesn’t offer any further details on what content to expect. However, the very nature of the event points a clear arrow at AMD’s forthcoming Instinct Mi300 accelerator. MI300 is AMD’s first shot at building a true data center/HPC-class APU, combining the best of AMD’s CPU and GPU technologies. AMD has offered only a handful of technical details about MI300 thus far – we know it’s a disaggregated design, using multiple chiplets built on TSMC’s 5nm process, and using 3D die stacking to place them over a base die – and with MI300 slated to ship this year, AMD will need to fill in the blanks as the product gets closer to launch.

As we noted in last week’s AMD earnings report, AMD’s major investors have been waiting with baited breath for additional details on the accelerator. Simply put, investors are treating data center AI accelerators as the next major growth opportunity for high-performance silicon – eyeing the high margins these products have afforded over at NVIDIA and other AI-adjacent rivals – so there is a lot of pressure on AMD to claim a slice of what’s expected to be a highly profitable pie. MI300 is a product that has been in the works for years, so the pressure is more of a reaction to the money than the silicon itself, but still, MI300 is expected to be AMD’s best opportunity yet to capture a meaningful portion of the data center GPU market.

MI300 aside, given the dual AI and data center focus of the event, this is also where we’re likely to see more details on AMD’s forthcoming EPYC “Genoa-X” CPUs. The L3 V-Cache-equipped version of AMD’s current-generation EPYC 9004 series Genoa CPUs, Genoa-X has been on AMD’s roadmap for a while. And with their consumer equivalent parts already shipping (Ryzen 7000X3D), AMD should be nearing completion of the EPYC parts. AMD has previously confirmed that Genoa-X will ship with up to 96 CPU cores, with over 1GB in total L3 cache available on the chip to further boost performance on workloads that benefit from the extra cache.

AMD’s ultra-dense EPYC Bergamo chip is also in the pipeline, though given the high-performance aspects of the presentation, it’s a little more questionable whether it will be at the show. Based on AMD’s compacted Zen4c architecture, Bergamo is aimed at cloud service providers who need lots of cores to split up amongst customers, with up to 128 CPU cores on a single Bergamo chip. Like Genoa-X, Bergamo is slated to launch this year, so further details about it should come to light sooner than later.

But whatever AMD does (or doesn’t) show at their event, we’ll find out on June 13^th at 10am PT (17:00 UTC). AMD will be live streaming the event from their website as well as YouTube.

Unlike other makers of cooling systems, Noctua has its roadmap advertised on its websites and always updates it to reflect changes in its product development plans. The company's May 2023 roadmap brings several surprises as it adds 'Next-gen AMD Threadripper coolers' and removes white fans from its plans.

The main thing that strikes the eye in Noctua's roadmap is the mention of 'next-gen AMD Threadripper coolers' coming in the third quarter. These products were not on the roadmap in January, per a slide published by Tom's Hardware. AMD has been rumored to introduce its next-generation Ryzen Threadripper processors for workstations for a while, but this is almost the first time when we have seen a more or less official confirmation about the existence of such plans, albeit not from AMD, but one of its partners. 

Since the confirmation does not come from the CPU developer, we would not put our money into launching the next-generation Ryzen Threadripper based on the Zen 4 microarchitecture in Q3. Meanwhile, it is reasonable to expect AMD's codenamed Storm Peak processor to arrive sooner than later since the company has not updated this lineup in a while.

Other notable things in Noctua's roadmap are a bunch of Chromax black products due in Q4, a 24V to 12V voltage converted set to arrive in Q2, and a 24V 40-mm fan, which emphasizes that the company considers the ATX12VO ecosystem essential to address. In addition, the firm is prepping its next-generation 140-mm fans, which will arrive in Q1 2024 in regular colors and then later in the year in Chromax—black version.

Unfortunately, Noctua's next-generation NH-D15 cooler, which once was promised to arrive in Q1 2023, is not slated for sometime in 2024. Meanwhile, the company's roadmap no longer includes white fans for a reason we cannot explain. Perhaps, the company decided to devote its resources elsewhere, or maybe white plastic that the company considered for white fans did not meet its expectations.

Source: Noctua

The head of Samsung's semiconductor unit acknowledged last week that the company's current mass production, leading-edge process technologies are a couple of years behind TSMC's most advanced production nodes. But Samsung is working hard to catch up with its larger rival in five years. 

"To be honest, Samsung Electronics' foundry technology lags behind TSMC," said Dr. Kye Hyun Kyung, the head of the Samsung Electronics Device Solutions Division, overseeing global operations of the Memory, System LSI and Foundry business units," at a lecture at the Korea Advanced Institute of Science & Technology (KAIST), reports Hankyung. "We can outperform TSMC within five years."

Samsung has been investing tens of billions of dollars in its foundry division in the recent years in a bid to catch up with TSMC and Intel, both in terms of production capacity for LSI chips as well as process technology advantages. The company has significantly closed the gap with its rivals, but it is still not quite on par with TSMC's fabrication technologies when it comes to performance, power, area (transistor density), and cost metrics.

While Samsung Foundry is the first contract maker of chips to adopt gate-all-around (GAA) transistors with its SF3E (3GAE, 3 nm, gate-all-around early) node, and the company's customers are enthusiastic about the technology itself and the novel transistor architecture, this process is not used for Samsung's own leading-edge system-on-chips for smartphones. 

"Customers' response to Samsung Electronics' 3nm GAA process is good," said Dr. Kye Hyun Kyung.

Meanwhile, Samsung's latest Galaxy S23-series uses Qualcomm's Snapdragon 8 Gen 2 SoC is made by TSMC on its N4 fabrication process.

Samsung Foundry's most advanced technology that can be used to make highly-complex SoCs for smartphones or other demanding applications is SF4 (4LPP, 4 nm, low-power plus), which, as the company admits, is significantly behind TSMC's N3 (N3B) node, is rumored to be used for mass production of Apple's highly-complex SoCs at this time.

The company may somewhat close the gap with TSMC's N3 and N4P with its SF4P (4LPP+) that will be available for customers later this year, according to a clarification published by @Tech_Reve.

Samsung Foundry will have a better chance to catch up with TSMC when its SF3 (3GAP) fabrication node enters high volume production in 2024, though by the time TSMC will also be offering its more advanced N3P manufacturing technology.  Around the same time Samsung also plans to offer SF4X (4HPC), a 4 nm-class fabrication technology that will (as the name suggests) address high-performance CPUs and GPUs.

Samsung reportedly believes that transition to GAA transistors in the 2022 ~ 2023 timeframe makes a great sense since it will have time to fix teething problems of the new architecture ahead of its rivals, most notably Intel and TSMC. As a result, when they start fabbing chips on their 2 nm-class technologies (20A, N2) in 2024 – 2025 and possibly encounter the same issues that Samsung is solving today, its SF2 node will be able to offer a better combination of power, performance, transistor density, costs, and yields.

Source: Hankyung.com (via @Tech_Reve)

At a recent OCP Regional Summit held in Prague, AMD shared its plans to replace its AMD Generic Encapsulated Software Architecture (AGESA) firmware with an open-source alternative called Open-Source Silicon Initialization Library (openSIL). The new firmware would be ready for production use in 2026, following a multi-year, four phase development cycle.

Firmware is a crucial component for modern computer systems, and on modern AMD systems, that critical code blob is AGESA. Among other things, AGESA is responsible for initializing several sub-systems of the platform, including processor cores, chipset, and memory; and it is frequently updated to support new hardware and resolve bugs.

But for all the utility that firmware brings, it can also be a weak point in a system via vulnerable to cyber attacks. So as part of their new firmware initiative, AMD has proposed making the development, architecture, and validation of the Silicon Initialization Firmware open-source to enhance security. AMD has a history of supporting open-source solutions for software and drivers, and openSIL is designed to be lightweight, simple, transparent, and secure and can be scaled easily.

Image Credit: AMD 

As initially reported on by Phoronix, openSIL is not intended to replace the Unified Extensible Firmware Interface (UEFI) but rather to be integrated with other host firmware such as coreboot, reboot, and FortiBIOS. It is written in standard industry language, allowing vendors to statically link it to the host firmware and bypass any host firmware protocols.

AMD is presently testing openSIL in the Proof-of-Concept (POC) phase, and it is currently compatible with AMD's 4th-generation EPYC (Genoa) processors and related platforms. The 5th-generation EPYC (Turin) processors will also be included in the POC phase. AMD intends to make openSIL the default choice for the 6th-generation EPYC series by 2026, and AGESA will be phased out.

Image Credit: AMD

While AMD admits that openSIL is still a work in progress, it is very close to parity with AGESA. However, since openSIL won't be ready until 2026 and AMD's most recent roadmap shows Zen 5 for 2024, it may take until Zen 6 or even Zen 7 before seeing a finished product. AMD has not released a projected roadmap for openSIL on the client side, but it will eventually replace AGESA on all AMD products.

Gallery: AMD openSIL OCP Presentation Deck

Source: AMD (via Phoronix)

One of the perks of modern mobile CPUs is that, being designed for laptops and their very limited cooling capacity, they can be placed in a NUC-sized (or smaller) miniature desktop PC with little difficulty. Better still, with desktops providing room for proper heatsinks (i.e. fins), even passively cooled mini-PCs using laptop-grade silicon are more than viable. The only real drawback to these mini-PCs has largely been the niche nature of the market – leading to high prices and limited choices for higher-performing systems – which is why Topgro has been turning heads as of late with the release of their aggressively priced Intel 12th Gen Core-based K3 Mini PC.

Topgro is not a household name, but it sells a collection of compact desktop PCs at Amazon, including inexpensive machines for office and neat gaming systems. As discovered by FanlessTech, Topgro's K3 is the latest addition to the company's lineup, offering a passively cooled mini-PC system based around Intel's mobile 12^thGeneration Core 'Alder Lake' processors with Iris Xe integrated GPU with 96 EUs.

Recently posted on Amazon, Topgro had initially listed a complete Core i7-1255U system that shipped with 512GB of solid state storage and 16GB of DDR4 memory for just $369 (after discount), a dirt-cheap price for a Core-based mini-PC that is complete and usable out of the box. Though in a sign that Topgro may have been a bit too aggressive with their new PC, the price already went up by $100 to $469 $569 just in the time it took to finish writing this article.

Measuring 174 mm × 128 mm × 45 mm (6.85 × 5 × 1.77 inches), Topgro's K3 Mini PC is a rather compact machine. And since Alder Lake CPUs for notebooks are heavily packed with features, these K3 machines are quite capable. The small form-factor PC not only comes with a 96 EU integrated Intel Xe-LP GPU, it supports up to 64 GB of DDR4 memory using two SO-DIMM modules, two M.2-2280 SSDs (with a PCIe 4.0 and a PCIe 3.0 interface), and can house one 2.5-inch SATA HDD or SSD for bulk storage. Even Thunderbolt 4 is supported, owing to the fact that it's natively baked into Intel's mobile CPUs.

As for connectivity, Topgro's K3 provides everything that Intel's 12^th Gen Core platform for laptops has to offer and then some. This includes Wi-Fi 6 (enabled by Intel's AX200 module), two 2.5GbE ports (making the systems plausible for corporate environments), the aforementioned Thunderbolt 4-capable USB Type-C connector, four display outputs (DP 1.4, two HDMI 2.0, USB-C), and six USB Type-A ports (three USB 3.0, two USB 2.0), and audio jacks.

As noted earlier, arguably the most notable aspect of this PC is Topgro's aggressive pricing, especially given that it's a fanless machine. The sole K3 configuration Topgro is offering pairs Intel's Core i7-1255U (2P + 8E cores, 12 threads, 4.70 GHz, 12 MB L3, Iris Xe GPU with 96 EUs) with 16 GB of DDR4-3200 and a 512 GB NVMe SSD, with Windows 11 pre-loaded. The manufacturer is offering a $30 digital coupon on top of a (now) $499 base price, bringing the final price of the sytem down to $469 $569.

Update 5:00pm ET: It was apparently too good of a price to last. Despite starting at $369 less than a day ago, Topgro's K3 is now $569, a quick $200 hike that puts it on par with other Alder Lake-U systems.

Coincidentally or not, $469 is also Intel's list price for the Core i7-1255U. And while system vendors rarely pay the listed price – especially over a year after the CPU has launched – systems such as the K3 underscore how aggressive PC vendors are needing to be in order to move PCs amidst the current slump in the market. Coupled with DRAM and NAND prices that are bottoming out at record lows, it's increasingly becoming possible to find decent systems at a low price.

And while this is the only fanless model in Topgro's profile, the company is also offering actively cooled mini-PCs in a similar form factor, and with similarly aggressive pricing. A Core i9-12900H box with the same RAM and NAND runs for $679 after discounts; which although is a big step up from the K3 in terms of pricing, does net you Intel's top Alder Lake mobile CPU.

After a few teasers and months of waiting, Corsair has finally launched the MP700, the company's first PCIe 5.0 SSD. The MP700 aims to win enthusiasts over with its ample capacity and high-speed performance. With sequential speeds up to 10,000 MB/s, the MP700 is ready to compete with the best SSDs that are presently on the market.

The MP700 is a standard M.2 2280 drive that fits into the PCIe 5.0 x4 M.2 interface and supports the latest NVMe 2.0 protocol. Initially, Corsair had advertised the MP700 with a thick cooler but ultimately decided to commercialize the drive without one. However, that doesn't mean consumers should run the MP700 au naturel since the SSD will likely suffer thermal throttling. Therefore, the recommendation is to use the motherboard's integrated M.2 heatsink or an aftermarket M.2 SSD cooler with the MP700 to ensure optimal operation. In addition, the MP700 features a double-side design, so that's something to consider when purchasing a retail M.2 cooling solution.

The MP700 features the Phison PS5026-E26 PCIe 5.0 SSD controller and Micron 232-layer 3D TLC NAND. The SSD flaunts sequential read and write speeds up to 10,000 MB/s on the 2 TB model. On the other hand, the 1 TB model has slightly lower specifications, with 9,500 MB/s sequential reads and 8,500 MB/s sequential writes. Random performance on the top SKU escalates to 1.7 million IOPS reads, and 1.5 million IOPS writes. Phison's E26 controller supports NAND speeds up to 2,400 MT/s, so on paper it can hit numbers as high as 15,000 MB/s. The MP700, like some of its other rivals, possesses NAND that operates at 1,600 MT/s, limiting the PCIe 5.0 drive's actual performance numbers to 10 GB/s. The MP700 also packs a decent DRAM package; on the 2 TB version, the drive comes with 4 GB of LPDDR4 memory. which is twice as much memory as we usually see on a 2 TB SSD.

Given that this is a high-end PCIe 5.0 SSD, Corsair is understandably eager to promote its use with Microsoft's DirectStorage API for faster game loading. While there's no specific hardware for this on the SSD side of matters – these are still bulk data transfer devices at the end of the day – Corsair and other vendors are aiming to sell their high-end SSDs to gamers who want the absolute highest performance possible – or at least the smallest bottlenecks. Sadly, Forspoken is the only game right now that supports DirectStorage. As a result, the feature isn't going to be a differentiator on Corsair's MP700 until more titles arrive with support for it.

The MP700's flagship performance comes at a cost. The Phison E26-powered SSD is a demanding drive in terms of power. The rated power consumption of the MP700 is around 10 watts for the 1 TB SKU and 10.5 watts for the 2 TB model. It draws substantially more power than some PCIe 4.0 drives, such as Corsair's own MP600 1 TB PCIe 4.0 SSD, rated for 6.5 watts. Power equals heat, which is the reason why the MP700 is dependent on a cooler to hit its maximum potential, especially during prolonged workloads.

The MP700, in its 1 TB presentation, has an endurance of 700 TBW, whereas the 2 TB flavor is good for 1,400 TBW. However, the ratings align with the competition since most PCIe 5.0 drives currently on the market use the same combination of the Phison E26 controller and Micron 232-layer NAND. In either case, Corsair backs the MP700 with a five-year warranty.

Corsair sells the MP700 1 TB (CSSD-F1000GBMP700R2) for $169.99 and the MP700 2 TB (CSSD-F2000GBMP700R2) for $289.99. These are premium price tags, and it's the price that consumers will have to deal with when it comes to early adopter technology.

Many brands have announced their PCIe 5.0 offerings, but few have hit the retail market. CFD Gaming's PG5NFZ drives are hard to come by in the U.S. market. On the other hand, Gigabyte's Aorus Gen5 10000 SSD comes back in stock every once in a blue moon. Meanwhile, the MP700 is available on Corsair's website and authorized worldwide retailers and distributors.

Supermicro this week began to list the industry's first commercial servers based on Intel's Data Center GPU Max 'Ponte Vecchio' compute GPUs. The machines use Ponte Vecchio in add-in-board and OAM module form factors, aiming at high-performance computing and large-scale AI training.

Supermicro currently has two servers qualified for Intel's Ponte Vecchio compute GPUs: the SYS-421GE-TNRT machine that can house up to 10 Data Center GPU Max 1100 cards with 480GB of HBM2 memory (48GB per board) as well as the SYS-821PV-TNR that can accommodate up to eight Data Center GPU Max 1550 OAM modules with 1TB of HBM2 memory onboard (128GB per card) and combined performance of 6.7 BF16/FP16 PetaFLOPS at 4.8 kW.

Both machines are based on two of Intel's 4^th Generation Xeon Scalable 'Sapphire Rapids' processors that can be mated with up to 8TB of DDR5 memory using 32 256GB modules. As for storage, both machines have 24 2.5-inch hot-swap bays for U.2/SATA/SAS drives (8x 2.5-inch NVMe hybrid; 8x 2.5-inch NVMe dedicated), and the SYS-421GE-TNRT also has two M.2 slots for PCIe drives.

For now, Supermicro sells the SYS-421GE-TNRT with Nvidia's A100 80GB or H100 80GB boards, but it looks like if requested, it can install Intel's Data Center GPU Max 1100 AIBs instead: the machine is fully qualified to run them, so throwing them in and installing required software should not take too long.

Meanwhile, the SYS-821PV-TNR that can house up to eight Data Center GPU Max 1550 OAM modules is listed as 'coming soon.'

Supermicro is currently the only supplier of commercial machines for AI and HPC workloads that offers systems equipped with Intel's Data Center GPU Max cards and modules. However, it is reasonable to expect other leading suppliers of servers to start selling similar products shortly.

Although Supermicro offers Intel Ponte Vecchio-based servers, it does not disclose their prices, as it always happens with AI and HPC machines configured individually and sold in quantities.

Source: Supermicro (@SquashBionic)

Solidigm has been active in the PC client SSD market with the QLC-based P41 Plus and the traditional TLC-based P44 Pro for the high-end market. While introducing the P41 Plus, Solidigm had also talked about the implementation of a read cache using the pSLC segment of the drive. This required the installation of Solidigm Synergy software. Over the last few quarters, the company has been hard at work perfecting the Solidigm Synergy 2.0 software release, with the promise of delivering even better performance for real-world workloads (compared to performance at launch) when used with the P41 Plus and the P44 Pro. While most SSD vendors work at the hardware and firmware levels, Solidigm believes there is performance benefits to be exploited at the software / driver level also.

Solidigm's Synergy 2.0 has two components - a Synergy Toolkit, and a Synergy Driver. While the toolkit taps into the SMART support and Windows performance counters and supports a variety of SSDs, the Synergy Driver is obviously supported only on Solidigm's client SSDs.

The toolkit itself is similar to the ones released by almost every other SSD vendor for the purpose of storage monitoring and maintenance. It collates a bunch of features that are spread over multiple tools and may be useful as a one-stop shop for mainstream users. The more interesting component is actually the Solidigm Synergy Driver that operates at the kernel level. Currently, this driver works only with Windows 10 or 11. Solidigm claims much better performance with its custom driver, with the QLC-based drive seeing significant improvement.

The driver is able to achieve this performance improvement using three different aspects:

• Dynamic Queues
• Prefetch for QD1 accesses
• FastLane (host-managed caching)

Out of these three, FastLane is available only for the P41 Plus currently. This host-managed caching scheme was already discussed in detail in our launch coverage of the P41 Plus, though it didn't have the FastLane moniker at that time.

Read caching is of help only when the drive is not full enough to actually make the cache size too small to be of good use. Solidigm claims that this technique helps most in drives that are between 25% to 75% full

The Synergy Driver includes a feature that analyzes the storage trace of an application in real-time to detect predictable read operations. When the access queue depth is low and operations are sequential in nature, it is possible to predict the next access and prefetch it prior to the actual application request. This can increase responsiveness from an user experience perspective.

The driver allows up to 8 streams to be fetched, each with a buffer size of 512KB and maximum request size of 128KB. [Update: These numbers can obviously be changed by Solidigm in a future driver release]. The driver's prefetch behavior can be controlled via a registry parameter [Update: Solidigm does not intend this key to be accessed by end users].

On systems with a large number of CPU cores, Solidigm claims that their driver can do a better job at routing I/O requests to relatively idle cores compared to the native Windows driver. The driver taps into the I/O request load, request size, and I/O processing times with / without CPU core redirection to decide whether to activate the dynamic queues. This process is dynamic - if the driver detects that the redirection makes completion times worse, or there is a change in workload, the dynamic queueing behavior is dropped. Similar to the smart prefetching, this aspect can also be deactivated using the registry [Update: Solidigm does not intend this key to be accessed by end users].

Solidigm's approach to delivering value additions to their client SSD customers after purchase is a marked departure from other vendors who rely solely on firmware updates - mostly for bug fixes. Providing better performance over time with driver releases is welcome from the end-user viewpoint. A point to keep in mind here is that these features help with real-world workloads - and not for use-cases primarily dealing with large sequential transfers, and measured using ATTO or CrystalDiskMark.

Gallery: Solidigm Synergy 2.0 Amplifies P41 Plus and P44 Pro Performance with Custom Driver

Alongside AMD’s usual slate of financial figures as part of their quarterly earnings call, the company also offered a brief update on the state of the next-generation Ryzen Mobile 7040HS “Phoenix” CPUs. After initially being slated for a March arrival – and then pushed back to April – the laptops are finally launching. According to AMD, the first systems should start arriving at retailers in the next few weeks.

First detailed during AMD’s CES 2023 keynote, the Ryzen Mobile 7040HS series (codename Phoenix) is AMD’s first mobile-focused, monolithic die CPUs based on the Zen 4 architecture, and will be their flagship silicon for mobile devices for 2023. Besides incorporating AMD’s latest CPU architecture, Phoenix also adds into the mix an updated RDNA3 architecture iGPU, and for the first time in any AMD CPU, a dedicated AI processing block, which AMD has aptly named the Ryzen AI. All of which, in turn, is fabbed using TSMC’s 4nm process – making it the single most advanced piece of silicon out of AMD yet.

During a post-call Q&A with the press and analysts, AMD representatives offered an update on the state of 7040HS series processors, and the laptops they powered. According to AMD, the 7040HS series chips began ramping in Q1 – contributing to AMD’s client revenue for the quarter – and are now shipping to OEMs. As a result, AMD expects the first systems to hit retailers in the next couple of weeks.

If nothing else, the extended launch timeline underscores the difference between how AMD and arch rival Intel communicate product launches – and what they count as shipping. Whereas Intel’s massive client volume requires a large product ramp-up such that they’ll ship mobile chips to OEMs sometimes months before anything is announced, AMD has in recent years been satisfied to announce new mobile hardware well in advance of system availability. As a result, whether it’s intentional or not, most of the time AMD ends up defining a mobile launch as when chips are shipping to OEMs, rather than the availability of OEM systems. And that seems to be what has happened here.

For our part, back on April 30^th AnandTech did find a single US retail listing claiming to have a Ryzen 7040HS laptop in stock. However, that listing from EXCaliberPC, for an ASUS ROG Zephyrus G14, now reads as a pre-order, with the retailer expecting it in stock on May 5^th. Assuming they do receive systems on that day, then this would put the arrival of the very first systems a bit ahead of AMD’s latest projections – though still missing the previous claim of “OEM partners to launch the first notebooks powered by Ryzen 7040HS Series processors in April” by a week.

Ultimately, despite the delays in getting Phoenix out the door, AMD does have big plans for their first monolithic Zen 4 product. Besides serving as the cornerstone of their high-performance mobile offerings for the next year, Phoenix is also expected to be used in AMD's Ryzen Z1 processors for handheld PC game consoles. And an eventual Ryzen Embedded product using the silicon is practically obigatory. So Phoenix will indeed rise high at AMD, eventually.

Update 4:30pm ET: And it turns out AMD's partners will have done one better than that. B&H Photo has an ASUS TUF Gaming A15 in stock, which is a Ryzen 7940HS-based laptop. So the first Phoenix laptops have indeed arrived, and sooner than AMD's most recent projections.

Continuing our earnings season coverage for Q1’2023, today we have the yin to Intel’s yang, AMD. The number-two x86 chip and discrete GPU maker has been enjoying a growing diet of Intel’s lunch for much of the past two years, but like the rest of the tech industry, AMD is now seeing a significant drop-off in sales as businesses and consumers alike curtail tech spending. So, like numerous other companies in this field, AMD has been bracing themselves for a rough first half of the year, after managing to beat the pack in Q4’2022 by posting a small profit.

For the first quarter of 2023, AMD reported $5.4B in revenue, continuing the year-over-year slide they’ve been experiencing throughout the past few quarters. All told, AMD’s top-line revenue dropped by 9% versus Q1’22, which is a much smaller drop than some of its rivals, but it’s a situation that has also been artificially buoyed by the Xilinx acquisition. That acquisition, which closed late into Q1’22, has boosted AMD’s revenue significantly on a year-over-year basis for the past few quarters – suffice it to say, it was not a small acquisition – though that is coming to an end now that AMD has owned Xilinx for over a full year.

The additional revenue and sales that Xilinx has brought to the table has shifted AMD in some important ways, but it hasn’t helped to halt the fundamental drop in sales that the tech industry is facing. As a result, the hit to AMD’s income has been significant; operating incomes dropped by 115% to negative $145M, and net income saw a similar dive into the red with a 118% drop, for a final tally of a $139M loss.

In an unusual move, Intel's director of global communications, Bernard Fernandes, took to Twitter this morning to confirm that the chip giant is preparing some chip branding change for later this year. Citing that the company is at an “inflection point” ahead of the launch of their Meteor Lake architecture client CPUs later this year, Intel is apparently developing something new for branding their first mass-scale chiplet-based consumer CPU.

Yes, we are making brand changes as we’re at an inflection point in our client roadmap in preparation for the upcoming launch of our #MeteorLake processors. We will provide more details regarding these exciting changes in the coming weeks! #Intel

— Bernard Fernandes (@Bernard_P) May 1, 2023

While the tweet in question doesn’t specifically address what it’s in response to, from context and timing it’s almost certainly a reaction to recent rumors that Intel is preparing to change their branding strategy for their Core family of consumer chips. And while it’s AnandTech policy not to republish or otherwise comment on rumors, an official comment from a high-ranking Intel PR representative means that this is no longer a mere rumor, and that changes are indeed in the works.

The current Core paradigm has been in place since late 2008 with the launch of the generational Core family and the now familiar i3/i5/i7 tiers. While Intel has since added the i9 tier and played with suffixes a few times in the last 15 years, Core branding has remained relatively consistent as a whole for what’s become 13 generations of parts.

But if the rumors are true, then the Core family will soon lose its i-series moniker. Based on benchmark data uploaded to the Ashes of the Singularity benchmark results database – a notorious source of leaks from hardware testers who neglect to turn off results reporting – an Intel Core Ultra 5 1003H has been spotted in the database. Which in turn suggests that Intel is intending to phase out the Core i-series branding for new Core Ultra branding.

To be sure, Intel’s tweet does not confirm the Ultra branding; and Ashes alone is not an authoritative source. But given that Intel has opted to confirm that they are making some brand changes to align with the Meteor Lake launch, if it's not Ultra, then some other kind of branding change is clearly in the works.

Microsoft has confirmed that the company will discontinue selling PC accessories under the Microsoft brand, narrowing the firm's focus to premium-priced peripherals sold under the Surface brand. The decision marks a major fork in the road for Microsoft-badged keyboards and mice, an era that started in 1983. And while the company isn't going to cease the production of PC accessories entirely, the shift to Surface represents a much smaller scope in products going forward. This change, in turn, calls into question the future of ergonomic peripherals at Microsoft, a traditional niche for the company that has resulted in some of their best-known (and most beloved) PC accessories.

"Going forward, we are focusing on our Windows PC accessories portfolio under the Surface brand," Dan Laycock, senior communications manager at Microsoft, told The Verge. "We will continue to offer a range of Surface branded PC Accessories — including mice, keyboards, pens, docks, adaptive accessories, and more. Existing Microsoft branded PC accessories like mice, keyboards, and webcams will continue to be sold in existing markets at existing sell-in prices while supplies last."

The statement contradicts information published by Nikkei about Microsoft's alleged plans to scale back production of Surface-branded gadgets. 

"We were recently informed by the client [Microsoft] to stop making stand-alone keyboards," an executive at a Microsoft supplier reportedly told Nikkei. "We were told the Surface series will still be one of Microsoft's development focuses, but just not the peripherals anymore."​

Microsoft confirmation about its focus shift towards Surface-branded PC accessories comes several months after the company announced a change in its hardware portfolio, which was a part of some 10,000 job cuts. The PC market is struggling due to macroeconomic challenges and uncertainty among consumers. In fact, Microsoft's own devices revenue, which includes Surface, PC accessories, and HoloLens, declined by 30% year-over-year in its most recent quarter, The Verge notes.

Microsoft-branded gadgets have been rather popular on the market over the past four decades, and while behind suppliers like Logitech, HP, and Dell, they've held their place – especially in the ergonomics market. Still, it has been getting increasingly hard for Microsoft to maintain its market share given the increasing number of players in the PC peripherals segment in general. The lucrative gaming keyboards and mice industry has been dominated by companies like Razer and Logitech for a while and then relatively new entrants like Corsair have not been making life of Microsoft any easier in the past decade with their highly-competitive products.

Surface-branded peripherals are a different segment though. They are not aimed at gamers and are not aimed at general users. Instead, they are designed for consumers willing to pay extra for advanced experience and businesses/enterprises for specific functionality. In fact, even Microsoft's recently released Surface-badged Thunderbolt 4 hub is clearly designed for enterprise uses that need functionality like remote management.

Ultimately, although Microsoft is not exiting the peripherals market entirely, the company's change in plans for the future of their hardware accessories seems to be much more significant than just a branding exercise. While Microsoft-designed mice, keyboards, and other devices will live on, narrowing their focus to premium, Surface-branded parts is a big change for a market that Microsoft has been a part of since almost the dawn of the PC – and likely not one for the better.

While they've long since given up developing GPUs, Matrox has remained a notable player in the video card industry for over four decades. These days, the company has settled into a modest role of providing graphics cards based on other vendors' GPUs for use in niche or boutique use cases, where Matrox can differentiate based on their software and support. And while their sales volume as a whole is limited, there's a certain degree of validation that comes from Matrox tapping a vendor's GPUs for their latest video cards.

To that end, Intel this week has finally earned their tip of the hat from the oldest of the video card vendors, with the announcement of a new series of multi-monitor display cards built around Intel's Alchemist architecture GPUs. Matrox's new Luma series graphics cards are based on Intel's Arc A310 and A380 graphics hardware, with the niche video card maker looking to tap into the Alchemist's class-leading video decoding and encoding capabilities, as well as the display output features and flexibility that are critical for a multi-display card.

Matrox Video's Luma family of graphics boards includes three products: the full-sized single-slot Luma A380 based on the Arc A380 (ACM-G11 with 1024 stream processors) with 6GB of memory; the low-profile single-slot fanless Luma A310 based on the Arc A310 (ACM-G11 with 768 stream processors) with 4GB of memory; and the low-profile single-slot Luma A310F that is equipped with a active cooling system. 

All three Matrox Luma graphics cards have four DisplayPort 2.1 UHBR10 (40Gbps) outputs and thus can drive four 4Kp144/5Kp60 monitors (uncompressed, 4:4:4) or two 8Kp60 or 5Kp120 displays using two of such ports for each displays. As for features, they offer the same capabilities as other Intel Arc A310/A380-based offerings, including support for DirectX 12 Ultimate, OpenGL 4.6, Vulkan 1.3, and OpenCL 3.0 APIs, as well as encoding and decoding of H.264, H.265, VP9, and AV1 video streams. As an added bonus, they retain support for Intel's oneAPI for compute tasks, as well as the Intel Distribution of OpenVINO toolkit for AI development.

All of the Luma graphics cards consume no more than 75W and can be powered entirely via a PCIe slot without any auxiliary power connectors, and the A310-based offerings are intended to fit into the most compact PCs that are out there. Furthermore, their single-slot design means that upwards of several cards can be installed into a single desktop PC for systems that need to drive eight, 12, or more monitors.

Meanwhile, it's interesting to note that while Matrox has not announced the discontinuation of their previous generation cards for these product segments – the NVIDIA based D-series and AMD-based M-series – in terms of specifications these new Intel cards should supplant the older cards in every way. Intel's DisplayPort 2.1 capabilities are likely the driving factor given Matrox's intended niche, with NVIDIA in particular being boxed out by not including DisplayPort 2.1 functionality with their Ada Lovelace generation GPU architecture.

Matrox Video's Luma boards are aimed primarily at the medical, digital signage, control room, video wall, and industrial markets. The cards come with a base three-year warranty, which can be further extended and a guaranteed lifecycle of seven years, which is important for some of the markets that they are intended for. 

Other advantages Matrox's Luma board offer include support for Matrox's PowerDesk software developed to handle exotic multi-display configurations. 

Matrox did not announce prices of the boards, though given their orientation on commercial, professional, and industrial applications, they will be priced accordingly.

Kicking off our coverage of the first earnings season of the year for the tech industry, we as always start with Intel. The blue-hued blue-chip is the first out of the gate to report their results for the first quarter of 2023, with Intel picking up the pieces after a rough end to 2023, and a rather painful start to 2023. With revenue down on a yearly basis almost across the entire board thanks to a major, industry wide slump in client and server sales, Intel’s focus has been on battening down the hatches to weather this rough period, while preparing for an eventual (if modest) upturn in the market later this year.

For the first quarter of 2023, Intel booked $11.7B in revenue, a precipitous 36% drop from the year-ago quarter. As was the case in Q4, Intel is in the midst of a major industry slump, which has hit revenues hard and operating/net incomes even harder. Intel closed the quarter in the red on an operating income basis, losing $1.5B, and the company’s overall net loss was a staggering $2.8B on a GAAP basis.

As the final set of announcements from this week's North American Technology Symposium, TSMC closed out their fab roadmap updates with some fresh news on their automotive-centric processes. For their automotive customers, TSMC is gearing up to deliver a pair of new specialized process nodes for the market, N4AE and N3AE, which are aimed at delivering early versions of TSMC's forthcoming automotive-grade process nodes for use by clients who need to get an early start with design and/or production.

The automotive industry is notoriously conservative when it comes to chip manufacturing, as automotive products are expected to meet a much higher bar in terms of safety and reliability. For good reason, the automotive industry is traditionally a generation or so behind in terms of the silicon lithography process nodes used, as they wait for them to become truly proven processes. With that said, the trend towards software-defined vehicles is driving a quick ramp-up in demand for more performant processors – and more chips altogether – which has been spurring the industry to close the gap a bit and transition to newer nodes sooner.

In order to meet that expected demand, this week TSMC announced its Automotive Early (AE) fabrication processes, which are aimed at helping automakers start earlier on the design of new chips for leading-edge nodes. Essentially stepping-stone nodes along the path to developing a full, automotive-grade N3A process, TSMC's N3AE (3nm class) and N4AE (4nm class) technologies are slated to offer consumer-grade reliability and will come with automotive-specific process design kits (PDKs) based on N3E and N4P respectively.

For customers who need a traditional, high-reliability automotive-grade chip, N3AE can be used by chip designers later this year to start developing automotive grade products that, in turn, will be fabbed on the fully-qualified automotive-grade N3A production node in a couple of years. Alternatively, customers who want to bring  chips to the market sooner to drive less critical systems in a car – such as digital cockpits and other non-critical systems – can plan to go into production with N4AE. Production on an Automotive Early node means forgoing some of the benefits of a fully qualified automotive process node – essentially building chips that are closer to consumer grade – but it will allow chip designers to get their chips in production on cutting-edge nodes upwards of a year sooner.

Traditionally, automakers and developers of automotive-grade electronics demanded their chips to be qualified for quality and reliability both in terms of functional safety (classified by ASIL system) and physical robustness (classified by AEC-Q100 standards). Improving the physical robustness of chips requires process technologies and design rules that have been specifically developed/tweaked meet those higher requirements, and typically it takes foundries and their automotive chip partners two or three years to develop an automotive-grade process technology. For example, TSMC has been making chips for smartphones and PCs on its N5 (5nm-class) fabrication process since 2020, but automotive grade N5A is only finally set to be ready this year.

At the same time, not all systems inside a vehicle need to comply with the most stringent integrity and reliability standards. For example, chips powering infotainment, digital cockpit, and advanced driver assistance systems (ADAS) do not have to be AEC-Q100 Grade 1 (handle temperatures between -40°C and +150°C). These systems also tend to require a lot of processing horsepower, leading to them significantly benefiting from leading-edge nodes.

TSMC said that its N3AE (3nm Auto Early) will be launched in 2023 and will provide automotive process design kits (PDKs) based on N3E. It will enable customers to initiate designs on the 3nm node specifically for automotive applications, ultimately leading to the N3A process, which is expected to be fully qualified for automotive use in 2025.

Meanwhile, customers in more of a hurry will also have N4AE as an option. N4AE is based on TSMC's existing N4P process technology, and customers will be able to start risk production on N4AE a year sooner, in 2024. Presumably, given its heritage from N4P, N4AE would also be a good candidate for incorporating IP already designed for N4P (which there is already piles of), further reducing the time to market.

TSMC did not touch upon which of its customers are interested in using its N3AE and N4AE process technologies for their automotive SoCs, but there are several usual suspects that have been offering high-performance feature-rich automotive SoCs for years.

As part of today’s Q1’2023 earnings announcement from Intel, the company is offering a brief update on the state of its upcoming chips and manufacturing nodes. Key among these disclosures is a status update on Meteor Lake, Intel’s next-generation client processor (and first disaggregated chiplet CPU), as well as the Intel 4 process node that it will be built upon.

According to Intel, the company is on track to launch Meteor Lake in the second half of the year. And more specifically, at this point Meteor Lake is now in production and ramping alongside the new Intel 4 process.

The basis of what we expect will be sold as Intel’s 14^th generation of Core processors, Meteor Lake is a critical part for Intel that represents several firsts for the company. It’ll be the first client CPU using chiplets, but also the first heterogenous chiplet CPU from the company, using different chiplets for I/O, CPU cores, graphics, and more. And on the manufacturing side of matters, it’s going to be the lead product for Intel’s next-generation EUV-based Intel 4 process node, as well as Intel’s biggest use of Foveros 3D die stacking technology yet – and the first attempt to do so for a relatively cheap, mass-market processor.

Intel’s last significant update on Meteor Lake came almost exactly a year ago, when the company announced during their Q1’2022 earnings report that it had been successfully powered on for testing. So at a year later, Intel is now making final preparations for a product launch later this year. The official launch window is still in the second half of the year – Intel client processor ramp-ups take several months even without all-new manufacturing technologies in play – so we expect Intel will announce more concrete details on that launch some time in the third quarter.

The official statement from AMD is as follows:

To counteract the problem, AMD has apparently identified an issue with specific chip voltages going too high when users enable AMD's EXPO memory profiles. A new cap on SoC voltages looks to be the primary change in the AGESA firmware rollout.

However, AMD's broad statement mentions that the update will address multiple power rails, which implies to some degree that the issue may be more than just the SoC power rail – or at least, that AMD isn't taking any chances. So what this entirely means is still a bit up in the air, as AMD hasn't specified in detail what it's doing outside of SoC power limits to prevent Ryzen CPUs from exceeding their specification limits.

Technically speaking, enabling EXPO memory profiles is a form of overclocking – i.e. operating the processor outside of specifications – as AMD's Ryzen 7000 family only officially supports DDR5 memory up to DDR5-5200 speeds. So going past this is putting additional stress on the memory controller in terms of clockspeeds; but the greater concern is how the various voltages on the chip are being adjusted to keep up with the demands of higher memory speeds.

One interesting point about AMD's statement is that it doesn't allude to whether or not the issue is just on its Ryzen 7000X3D processors, or whether it affects all of its Ryzen 7000 processors entirely. Regardless of the Zen 4 chip that users may have, AMD is ambiguous in its language, and it seems to be that AMD is recommended that all users with a Ryzen 7000 series processor should update to the latest firmware.

In practice, enabling EXPO memory profiles on compatible DRAM does seem to push SoC voltages beyond AMD's safe spot on the Ryzen 7000 processors, which AMD is treating as part of the cause of the burnout issue. AMD does, however, state that the changes it has made to their AGESA firmware, once flashed, shouldn't affect the user's ability to apply EXPO memory profiles on compatible kits of DDR5 memory. Which does raise the question of why motherboards were increasing SoC voltages in the first place, as presumably this shouldn't be needed if AMD's new caps won't limit EXPO memory overclocking.

In any case, AMD is actively working with its motherboard partners to release a new AGESA firmware with the new voltage limits, which they say has already been distributed. AMD claims that all AM5 motherboard vendors and models should have a new BIOS version available to them within the next few days, and is recommending all users to update their BIOS at their earliest convenience.

Image source: Speedrookie/Reddit

Alongside some new announcements for their 2nm process node plans, TSMC has also released a progress and roadmap update for their N3 family process technologies at today's 2023 North American Technology Symposium. TSMC's final generation of FinFET-based process nodes, the N3 family is expected to remain around in some shape or form for many years to come as the densest node available for customers who don't need a more cutting-edge GAAFET-based process.

TSMC's big roadmap updates on the N3 front were with N3P and its high performance variant, N3X. As revealed today by TSMC, N3P will be an optical shrink of N3E, offering enhanced performance, reduced power consumption, and increased transistor density compared to N3E, all while maintaining compatibility with N3E's design rules. Meanwhile, N3X will wed extreme performance with 3nm-class densities, offering higher clockspeeds for high performance CPUs and other processors.

N3E: 3nm for Everyone Is on Schedule

As a quick refresher, TSMC's N3 (3nm-class) family of process technologies is comprised of several variants, including baseline N3 (aka N3B), relaxed N3E with reduced costs, N3P with enhanced performance and chip density, and N3X with higher voltage tolerances. Last year the company also talked about N3S with maximized transistor density, but this year the company remains tight lipped about this node and it is not mentioned anywhere in its slides.

TSMC's vanilla N3 node features up to 25 EUV layers (according to China Renaissance and SemiAnalysis), with TSMC using EUV double-patterning on some of them to make for higher logic and SRAM transistor density than N5. EUV steps are expensive in general, and EUV double patterning drives those costs up further, which is why this fabrication process is only expected to be used by a handful of customers who are not as concerned about the high outlay required.

The majority of TSMC's clients interested in a 3nm-class process are expected to use the relaxed N3E node, which according to TSMC is on schedule and achieving their performance targets. N3E uses up to 19 EUV layers and does not rely on EUV double patterning at all (according to data from China Renaissance and SemiAnalysis), reducing its complexity and costs. The trade-off is that N3E offers lower logic density than N3, and it has the same SRAM cell size as TSMC's N5 node, making it somewhat less attractive to those customers who are driving for density/area gains. Overall, N3E promises a wider process window and better yields, which are two crucial metrics in chip fabrication.

"N3E will be better than N3 in terms of yield, process complexity, which directly translating into [wider] process window," said Kevin Zhang, Vice President of Business Development at TSMC. 

N3P: Higher Performance, Higher Transistor Density

Following N3E, TSMC will continue to optimize the N3 family's transistor density with N3P, which will build on N3E by offering improved transistor characteristics. The refined process node will allow chip designers to either increase performance by 5% at the same leakage or reduce power by 5% ~ 10% at the same clocks. The new node will also increase transistor density by 4% for a 'mixed' chip design, which TSMC defines as a chip consisting of 50% logic, 30% SRAM, and 20% analog circuits.

As part of their discussion on N3P, TSMC stressed that the density improvement was achieved by adjusting the optical performance of its scanners. So it is likely that TSMC will be able to shrink all types of chip structures here, which will make N3P an attractive node for SRAM-intensive designs.

"N3P is a performance boost, it has a 5% higher performance, at least a 5% higher than N3E," explained Zhang. It also has the 2% optical shrink, which bring transistor density to 1.04x."

Since N3P is an optical shrink of N3E, it will preserve N3E's design rules, enabling chip designers to quickly reuse N3E IP on the new node. Consequently, N3P is also projected to be one of TSMC's most popular N3 nodes, so expect IP design houses like Cadence and Synopsys to offer a variety of IP for this process technology, reaping the benefits of forward compatibility with existing N3E in the process. TSMC says that N3P will be production ready in the second half of 2024.

N3X: Maximum Performance and Density

Finally, for developers of high-performance computing applications like CPUs and GPUs, TSMC has for the past few generations been offering its X family of high voltage, performance-focused nodes. As disclosed at last year's event, the N3 family will be getting its own X variant with the aptly named N3X node.

Compared to N3E, N3X is projected to offer at least 5% higher clockspeeds compared to N3P. This is being accomplished by making the node more tolerant of higher voltages, allowing chip designers to crank up the clockspeeds in exchange for higher overall leakage.

TSMC claims that N3X will support voltages of (at least) 1.2v, which is a fairly extreme voltage for a 3nm-class fabrication process. The leakage cost, in turn, is significant, with TSMC projecting a whopping 250% increase in power leakage over the more balanced N3P node. This underscores why N3X is really only usable for HPC-class processors, and that chip designers will need to take extra care to keep their most powerful (and power-hungry) chips in check.

As for transistor density, N3X will offer the same density as N3P. TSMC has not commented on whether it will also maintain design rule compatibility with N3P and N3E, so it will be interesting to see what ends up happening there.

The final N3 family node in TSMC's current roadmap, the company says that that N3X will be production ready in 2025.

Memory module producers have been shipping unbuffered DDR5 memory modules for desktop and laptop computers running Intel's 12^th Generation Core 'Alder Lake' processors in high volumes since September, 2021, without any major issues. But DDR5 is just now entering the datacenter world, and according to a recent report, it looks like power management ICs (PMICs) for registered DIMMs have become a constraining factor due to compatibility issues.

In a report published by TrendForce discussing the current state of the market for server-grade DDR5 memory, the semiconductor analyst firm noted that there is an issue with PMIC compatibility for DDR5 RDIMMs, with both DRAM suppliers and PMIC vendors are collaborating to resolve the problem. The analysts do not reveal the exact root cause of the problem, but claim that PMICs from Monolithic Power Systems (MPS) do not have any issues, leading them to expect MPS PMICs to be in high demand for the foreseeable future.

Although DRAM makers have been distributing samples of their server grade modules to CPU and server makers since early 2022, practical issues only emerged recently when producers began to ramp up production of their machines running AMD's EPYC 9004 'Genoa' and Intel's 4^th Generation Xeon Scalable 'Sapphire Rapids' processors. As a result, the demand for PMICs from a single supplier has created a bottleneck in production, claims TrendForce. This will have a knock-on effect on the server market, which is already suffering from a demand drop.

Neither analysts nor DRAM producers are currently disclosing the precise reason for the PMIC issue. But it is evident that, as both client and server DDR5 DIMMs require PMICs, it is turning out to be harder to make server-grade modules than client-aimed DIMMs.

As part of the changes that came with the DDR5 specification, DDR5 memory modules now come with their own voltage regulating modules (VRMs) and PMIC. Moving these components on to DIMMs is intended to minimize voltage fluctuation ranges (DDR5's allowable range is about 3% (±0.033V) for a 1.1 volt supply), as well as decrease power consumption and improve performance. But doing so adds complexity to individual DIMMs, as well. 

Unbuffered DDR5 DIMMs for client PCs are relatively simple since they are all single or dual-rank and carry at most 16 single-die memory chips. High-capacity Registered DDR5 memory modules for servers use more chips and those chips can pack in multiple DRAM dies each, which greatly increases complexity.

As a result of the PMIC bottleneck as well as a slower ramping of DDR5 manufacturing capacity, TrendForce predicts that prices of server-grade 32GB DDR5 modules will drop to around $80 - $90 in April and May, due to the lower fulfillment rates of DDR5 server DRAM in the short term. As a result, DDR5 prices are expected to fall more slowly than DDR4 for the next couple of quarters, with DDR5 prices only finally catching up (or rather, down) with DDR4 once production picks up.

Corsair has introduced its first ever SSD in an M.2-2230 form-factor that is aimed at ultra-compact systems like Valve's Steam Deck as well as Microsoft's Surface Pro 8 and 9. The MP600 Mini drive delivers 1 TB of storage space and strong performance in a miniature package.

Corsair's MP600 Mini is based around Phison's E21T controller mated with 3D TLC NAND memory (according to Tom's Hardware) and is rated with an up to 4,800 MB/s sequential read/write speed as well as an up to 850K/1.1M random read/write IOPS. The SSD consumes about 4.3W on average and comes with a simplistic graphene heatspreader, which makes it compatible with the vast majority of systems out there.

Since Corsair's MP600 Mini uses a PCIe 4.0 x4 interface, it will shine in the vast majority of modern laptops and hybrid PCs. Though as far as Valve's Steam Deck is concerned, keep in mind that it uses a PCIe 3.0 x4 interface, so the MP600 Mini won't get to fully stretch its legs there in terms of throughpt.

The market for retail M2.2230 SSDs has traditionally been a small one, owing to the limited use of the drives in user-upgradable machines. However the recent launch of the Steam Deck in particular has kicked up significant demand for these SSDs – especially high-performance drives befitting a game console. So Corsair's MP600 Mini is a welcome addition to the space. 

Corsair currently sells its MP600 Mini directly for $110, which might seem expensive for a 1 TB drive, but keeping in mind that we are dealing with a pretty fast model that comes in an M.2-2230 form-factor (a rare combination), the price is consistent with similar drives.

AMD has announced the Ryzen Z1 series of high-performance processors for handheld PC gaming consoles. The latest Ryzen Z1 chips wield Zen 4 cores and RDNA 3 graphics, a familiar combination that AMD uses for its current mobile Ryzen 7040HS series (Phoenix) processors. And while the chipmaker hasn't officially announced any lower power Phoenix chips yet (e.g. Ryzen 7040U series), given the similarities in the feature set of the Ryzen Z1 family and the Phoenix silicon, the Ryzen Z1 series appears to be a custom-tailored Phoenix SKU that prioritizes energy efficiency for handheld gaming devices. The lead customer for the Ryzen Z1 will be Asus, who will be using it in their upcoming ROG Alloy.

The Ryzen Z1 series is being introduced with two SKUs. The top-tier Ryzen Z1 Extreme has eight Zen 4 CPU cores – the maximum found on Phoenix – while the vanilla Ryzen Z1 cuts that down to six CPU cores. Meanwhile, on the graphics side of matters, the Ryzen Z1 Extreme flaunts 12 RDNA 3 architecture compute units, again aligning with the max number found on Phoenix silicon. Its vanilla Z1 counterpart, on the other hand, will ship with just 4 CUs enabled, one-third as many as on the flagship part.

Both chips benefit from a 16 MB L3 cache, while the Ryzen Z1 Extreme has 2 MB more L2 cache due to its additional CPU cores. In any event, the Ryzen 1 and Ryzen Z1 Extreme possess strong computing performance thanks to the high CPU core count and the usage of AMD's latest Zen 4 cores. A hexa-core processor should suffice for handheld gaming, and either Ryzen Z1 processor will certainly outperform AMD's older custom quad-core Zen 2 SoC (Aerith) that powers Valve's Steam Deck.

Hardware unit counts aside, AMD isn't revealing the clock speeds for the two 4nm processors. Curiously, an embargoed report published by The Verge has quoted "up to" 8.6 TFLOPS and 2.6 TFLOPS graphics performance, respectively, based on AMD's engineering projections. But these figures have been removed from the deck AMD has since been sending around to the rest of the press. If these figures do turn out to be reasonably close to what the final shipping silicon can hit, then that would mean we'd be looking at peak GPU clock speeds of around 2.8 GHz and 2.6 GHz, respectively, which is just a bit lower than what the Ryzen 7040HS chips run at.

The Ryzen Z1 series chips will support a maximum TDP of 30 Watts, which happens to be the same top power target as AMD's mobile Ryzen U-series chips. Meanwhile, although this wasn't in AMD's press deck or press release, AMD's Ryzen Z1 product page lists a minimum TDP of 15 Watts. Though it's worth noting that, unlike TDP maximums, TDP minimums are not a hard floor or ceiling for mobile processors. Chips can be configured to run even less when it's called for, depending on what the customer desires.

Just to offer a comparison here, Aerith's TDP is between 4 Watts and 15 Watts, meaning the Z1 series certainly offers a good deal larger power range for customers to play with. At its top TDP, such a configuration would quickly blow through the battery in any handheld device, but for docked devices, it's very feasible. As for handheld mode, it's far more likely that we'll see Z1 chips clocked relatively low to conserve precious power.

Being that this is an AMD product designed for PC-like devices, it has access to all AMD graphics technologies, including Radeon Chill, which helps improve battery life by adjusting frame rates based on the movements in the game. This also means that FSR will be on the table on a game-by-game basis, and RSR as a less optimal universal solution.

Gallery: AMD Ryzen Z1 Series Slide Deck

AMD provided some benchmarks for the Ryzen Z1 series utilizing the Asus ROG Ally. As usual, throw a pinch of salt over vendor-provided benchmarks. The device used here also had 16 GB of LPDDR5 memory and a Micron 500 MB SSD. It's important to highlight that AMD put the Asus ROG Alloy in turbo mode for its benchmarks, which means that the Ryzen Z1 series processors were operating at power levels of up to 30 Watts.

AMD's results show the Ryzen Z1 Extreme delivers average frame rates above 60 FPS in most titles. However, the chipmaker tested with the lowest settings at a mix of 1080p and 720p upscaled with RSR to hit those figures. The Ryzen Z1's performance wasn't bad, but the two chips' delta was perceptible. The difference spans between 6% and 73%. Zen 4-based processors took a significant performance hit at native 1080p resolution on the lowest settings. As expected, the Ryzen Z1 struggled in multiple titles and didn't hit the 60 FPS mark. 

Asus has first dibs on the Ryzen Z1 series; however, more competitors powered by the 4nm Ryzen Z1 chips should arrive down the road. Asus' ROG Ally will be the first handheld gaming console to feature AMD's Ryzen Z1 Extreme processor, with the company set to share more information on the handheld (including pricing) on May 11.

Yesterday we reported that MSI announced a wave of firmware updates designed to address and alleviate potential issues with users on AM5 using AMD's Ryzen 7000X3D processors with 3D V-Cache. One of the main changes with MSI's latest UEFI firmware for AM5 included voltage restrictions when using Ryzen 7000X3D series CPUs so that these chips couldn't be overvolted as the V-Cache packaging is somewhat sensitive to additional power.

Further to MSI's announcement, ASUS has released a statement to experienced engineer and extreme overclocker Roman 'Der8auer' Hartung, which addresses the potential issue with using AMD's EXPO memory profiles in conjunction with the Ryzen 7000X3D series chips. One of the key elements that seemingly surrounds the problem is the use of AMD's EXPO memory overclocking profiles.

ASUS's Director of Global Product Marketing and Technical Marketing, Rajinder Gill, said in a statement to Der8auer, "The EFI updates posted on Friday contain some dedicated thermal monitoring mechanisms we've implemented to help protect the boards and CPUs. We removed older BIOSes for that reason and also because manual Vcore control was available on previous builds." Rajinder also said, "We're also working with AMD on defining new rules for AMD Expo and SoC voltage. We'll issue new updates for that ASAP. Please bear with us."

ASUS ROG Crosshair X670E Hero AM5 Motherboard for AMD Ryzen 7000 Series CPUs

Looking at the narrative of the statement provided to Der8uaer from Rajinder, we can confirm that ASUS has indeed removed older iterations of its firmware for its AM5 motherboards. Looking at the EMEA side of things, at the time of writing, the latest firmware update to be posted on the ASUS product pages is dated 04/21/2023, which indicates ASUS recently updated the firmware, but the update isn't the latest one in question.

Roman 'Der8auer' Hartung's Ryzen 9 7900X also has a bulge.

The one thing these issues have in common is that AMD's EXPO memory profiles have been applied, claims Roman 'Der8auer' Hartung, and even highlighted that his Ryzen 9 7900X processor also fell foul the the 'bulging issue.' This is particularly interesting as this isn't an X3D series chip with 3D V-Cache, although we're not aware of any other reports of non-X3D Ryzen 7000 chips being affected at this time.

Enabling EXPO memory profiles on Ryzen 7000 processors does several things to the processor that pushes it beyond the technical specification of the chip. Chief among these is raising the SoC voltage and some other primary voltages, such as VDDIO, which are used to feed the IOD. Like any ASIC, there are limits to how high voltages can be safely pushed, and this is one theory behind the cause of the recent run of damaged Ryzen reports. Though how this might be connected to the issue being centered around Ryzen X3D chips – where the voltage-sensitive V-Cache is on the CCDs and their separate voltage plane – is unclear at this time.

In any case, this does underscore why AMD's EXPO memory overclocking profiles void the warranty on these chips, as there's more to EXPO than just ramping up memory frequencies and applying more voltage to the DRAM itself.

Further to Rajinder's statement given to Der8auer, ASUS's Senior Technical Marketing Manager, Juan Jose Guerrero, posted the following statement on Twitter Tuesday afternoon:

As we mentioned in our article yesterday about MSI addressing CPU voltage on their own AM5 motherboards, vendors are looking to address any more potential issues, following a small-but-concerning number of reports circulating on Reddit that their Ryzen 7000X3D processors are burning out and killing the motherboard in the process. ASUS's statement and the summary of the firmware on the official product page indicate that ASUS is worried about SoC voltage, and as such, the new firmware locks it down to 1.30 V. Which ASUS states is to 'protect the CPU and motherboard.' 

Screenshot of the ROG Crosshair X670E Extreme BIOS/Support page on the official product page (China).

On the Chinese product page for the ROG Crosshair X670E Extreme, the firmware has been updated as of today (at the time of writing) to 04/25/2023, and the firmware version in question (1302) does seemingly address the SoC voltage.

Further to ASUS's and MSI's statements, AMD has officially responded with a statement of their own:

It reads, "We are aware of a limited number of reports online claiming that excess voltage while overclocking may have damaged the motherboard socket and pin pads. We are actively investigating the situation and are working with our ODM partners to ensure voltages applied to Ryzen 7000X3D CPUs via motherboard BIOS settings are within product specifications. Anyone whose CPU may have been impacted by this issue should contact AMD customer support."

Of course, these issues of Ryzen 7000X3D series processors burning out aren't just limited to those using MSI and ASUS motherboards. Since the initial reports, users have reported additional issues on various models and brands across threads on Reddit. The potential for damage is one of the primary reasons AMD locked down the X3D series processors so that users couldn't manually overclock them. Even the first iteration, the Ryzen 7 5800X3D, was also locked down.

One thing remains clear: AMD and its motherboard partners are now officially investigating the matter, and users with affected Ryzen 7000 CPUs are advised to contact AMD customer support directly.

The statement directly from AMD is as follows:

As the statement clarifies, AMD themselves are investigating the issue that users have been experiencing, which has been reported on through various channels such as social media and Reddit. This is further to MSI, which launched new firmware yesterday, and ASUS announced new firmware today for users with AMD Ryzen 7000X3D processors, such as the Ryzen 9 7950X3D and the Ryzen 7 7800X3D. The new firmware specifically limits the SoC voltages applied so that these critical limits aren't breached when users enable AMD EXPO memory profiles on supported DRAM.

While AMD hasn't officially confirmed the problem regarding the burnt pads on the Ryzen 7000X3D series processors and the burnt pins within the motherboard socket, AMD is actively looking to resolve this issue. One theory thus far is that the issue is being triggered when users are enabling AMD EXPO memory profiles – which, much like Intel's XMP counterpart, is technically a form of overclocking and officially voids the product warranty. Given the rollout of new firmware, which targets SoC voltages and restricts them, all things point towards voltages and overclocking said processors, which are, by default, multiplier and CPU VCore locked.

AMD is actively working with motherboard vendors such as MSI, ASUS, GIGABYTE, and ASRock to roll out the new firmware. AMD also specifies that if users have a Ryzen 7000X3D processor affected by this problem, to contact AMD customer support directly.

Update (04/27): AMD has officially made a second statement regarding the Ryzen 7000 and 7000X3D burnout issues. It is as follows:

Image source: Speedrookie/Reddit