NVIDIA has today announced its Grace CPU Superchip, a monstrous design focused on heavy HPC and AI processing workloads. Previously, team green has teased an in-house developed CPU that is supposed to go into servers and create an entirely new segment for the company. Today, we got a more detailed look at the plan with the Grace CPU Superchip. The Superchip package represents a package of two Grace processors, each containing 72 cores. These cores are based on Arm v9 in structure set architecture iteration and two CPUs total for 144 cores in the Superchip module. These cores are surrounded by a now unknown amount of LPDDR5x with ECC memory, running at 1 TB/s total bandwidth.

NVIDIA Grace CPU Superchip uses the NVLink-C2C cache coherent interconnect, which delivers 900 GB/s bandwidth, seven times more than the PCIe 5.0 protocol. The company targets two-fold performance per Watt improvement over today's CPUs and wants to bring efficiency and performance together. We have some preliminary benchmark information provided by NVIDIA. In the SPECrate2017_int_base integer benchmark, the Grace CPU Superchip scores over 740 points, which is just the simulation for now. This means that the performance target is not finalized yet, teasing a higher number in the future. The company expects to ship the Grace CPU Superchip in the first half of 2023, with an already supported ecosystem of software, including NVIDIA RTX, HPC, NVIDIA AI, and NVIDIA Omniverse software stacks and platforms.

KIOXIA America, Inc., today announced that it is sampling its 2nd generation PCIe 5.0 SSD for enterprise and data center customers. After being the first vendor to offer a drive using the PCIe 5.0 interface, KIOXIA has now introduced the new CD8 Series data center NVMe SSD family. CD8 SSDs double the bandwidth per lane over PCIe 4.0 SSDs from 16 gigatransfers per second (GT/s) to 32GT/s and are optimized for hyperscale data center and enterprise server-attached workloads.

"Today, PCIe 4.0 SSDs are considered at the forefront in terms of delivering the highest levels of drive performance," commented Greg Wong, founder and principal analyst, Forward Insights. "Next generation PCIe 5.0 SSDs provide twice the level of performance, and will continue to propel the PCIe/NVMe SSD market - which is expected to grow at a CAGR of over 20% out to 2026."

FSP Group, the leading manufacturer of high-performance power supplies today, announces product compliance with INTEL PSDG ATX 3.0 and PCIe 5.0 for the ultimate performance enhancement.

The retail models of FSP's consumer-based power supply range are getting a major overhaul with the addition of INTEL Power design guide ATX 3.0 and PCIe 5.0 specification compliance, packing more power and stability that is tailor-made for high demand systems. Traditional power supplies need three 8-pin to 16-pin adapters to support the latest PCIe Gen 5.0 graphics cards. FSP looks to address this drawback with their latest iteration of PSUs, unlocking seamless integration with all the latest hardware in the market.

Teledyne LeCroy, the worldwide leader in serial data test solutions is pleased to announce that the Peripheral Component Interconnect Special Interest Group (PCI-SIG), responsible for the definition of serial expansion buses and related components required to drive fast, efficient transfers between processors and peripheral devices, has approved Teledyne LeCroy as a solution vendor for official PCI Express (PCIe ) 5.0 link and transaction layer compliance testing at the PCI-SIG compliance workshops.

Compliance testing allows design and test engineers to ensure product interoperability as well as adherence to the PCIe standard, enabling technology adoption among integrators. The Summit Z58 Protocol Exerciser/Analyzer which was previously approved for PCIe 3.0 and 4.0 compliance testing has now also been approved for official PCIe 5.0 compliance and certification testing, making it the only piece of test equipment currently certified for link and transaction layer testing of all three PCIe generations. Passing the PCI-SIG's compliance tests are also a necessary step for equipment to be included in the PCI-SIG's Integrators List.

As data processing requirements continue to evolve, cloud-scale infrastructure requires solutions that maximize bandwidth, minimize latency and are optimized for efficient resource utilization. Microchip Technology Inc. today announced the newest member of the Flashtec controller family, the NVMe 4016 SSD controller. As the industry's highest-performing PCIe Gen 5 NVMe SSD controller, the Flashtec NVMe 4016 controller addresses the market demand for high-reliability, high-performance Solid-State Drives (SSDs) delivering greater than 14 GB per second throughput and over 3 million IOPS. With advanced Credit Management technology, this next-generation controller offers the stringent Quality of Service (QoS) required by today's cloud-focused data center applications.

"Microchip is proud to announce the next generation of our Flashtec NVMe controller product line. Its market-leading performance, coupled with our proven and flexible architecture, means the NVMe 4016 can provide our cloud and OEM customers with a cutting-edge platform for their PCIe Gen 5 NVMe SSD solutions," said Pete Hazen, vice president of Microchip's data center solutions business unit.

During the International Solid-State Circuits Conference (ISSCC) 2022, Intel gave us a more significant look at its upcoming Ponte Vecchio HPC accelerator and how it operates. So far, Intel convinced us that the company created Ponte Vecchio out of 47 tiles glued together in one package. However, the ISSCC presentation shows that the accelerator is structured rather interestingly. There are 63 tiles in total, where 16 are reserved for compute, eight are used for RAMBO cache, two are Foveros base tiles, two represent Xe-Link tiles, eight are HBM2E tiles, and EMIB connection takes up 11 tiles. This totals for about 47 tiles. However, an additional 16 thermal tiles used in Ponte Vecchio regulate the massive TDP output of this accelerator.

What is interesting is that Intel gave away details of the RAMBO cache. This novel SRAM technology uses four banks of 3.75 MB groups total of 15 MB per tile. They are connected to the fabric at 1.3 TB/s connection per chip. In contrast, compute tiles are connected at 2.6 TB/s speeds to the chip fabric. With eight RAMBO cache tiles, we get an additional 120 MB SRAM present. The base tile is a 646 mm² die manufactured in Intel 7 semiconductor process and contains 17 layers. It includes a memory controller, the Fully Integrated Voltage Regulators (FIVR), power management, 16-lane PCIe 5.0 connection, and CXL interface. The entire area of Ponte Vecchio is rather impressive, as 47 active tiles take up 2,330 mm², whereas when we include thermal dies, the total area jumps to 3,100 mm². And, of course, the entire package is much larger at 4,844 mm², connected to the system with 4,468 pins.

The NUC Extreme might not share much in terms of design with either the original NUC or its earlier Extreme siblings these days, as it's grown into a rather hefty SFF system. Now the 12th gen NUC Extreme has leaked and it comes with a few unexpected surprises, in both a good and a bad sense. Initially it looks like there will be two main barebones SKU's, the NUC12EDBi9 and the NUC12EDBi7, with a 65 W Core i9-12900 and a Core i7-12700 CPU respectively. In other words, it appears we're not looking at any dedicated CPU SKU's this time around.

The first thing that sticks out in the spec is the fact that Intel has gone for a pair of DDR4 SO-DIMM slots, rather than DDR5. In all fairness, this could be due to a lack of DDR5 SO-DIMMs in the market, but feels odd in a product with Extreme in the product name. Memory speeds of up to 3200 MHz are supported and up to 64 GB can be fitted. On the storage side, there's support for no less than three PCIe 4.0 M.2 NVMe drives using the 2280 form factor. Two of the slots can also accept 2242 drives and SATA drives. A full PCIe 5.0 x16 slot is also present for an optional graphics card, but more on that a little bit later.

Although we've already seen some companies both tease and announce PCIe 5.0 consumer SSDs, it seems like we shouldn't expect mainstream PCIe 5.0 SSDs until 2024, at least if Silicon Motion's earnings call is anything to go by. Wallace Kou, Silicon Motion's CEO was quoted as saying ""It is likely that PCIe Gen4 will last a few years since Intel, AMD both continue to bring new upgrade variant of CPU with PCIe Gen4 to the market," continuing "Similarly, we are preparing for the launch of our third-generation PCIe Gen4 controller next year before transitioning to PCIe Gen5 in the following year."

This obviously has something to do not only with development time, but also with recuperating the costs of developing the previous generation of PCIe 4.0 SSD controllers before introducing new products that would have limited market share due to at least initially high prices. Silicon Motion is said to be working on multiple PCIe 5.0 SSD controllers with the enthusiast grade SM2508 expected towards the end of this year and it's set to compete with Phison's PS5026-E26 and InnoGrit's IG5666 controllers, plus unannounced inhouse controllers from the likes of Samsung and WD. Silicon Motion's mainstream PCIe 5.0 controller is said to be the SM2507, which is also said to be more suitable for mobile applications, most likely due to lower power consumption. However, the first PCIe 5.0 controller from Silicon Motion should be the SM8366 enterprise grade controller that was announced last year and which should be shipping in the second half of this year. For most users, PCIe 5.0 SSDs are unlikely to make a huge difference, unless we see some major improvements when it comes to random read and write performance, as this is currently the real bottleneck with SSDs.

To meet the demand for high-speed storage driven by big data, TEAMGROUP is utilizing its years of experience in computer storage solutions to launch the newest PCIe Gen5 SSD this year. In light of the generational advancements of PCIe 5.0, the company is actively committed to developing products using cutting-edge technologies. With its expertise in R&D and dedication to product excellence, the company is excited to enter the next era of SSD storage solutions and to deliver spectacular new products to the world.

TEAMGROUP is releasing its first PCIe Gen5 SSD under its gaming sub-brand T-FORCE as part of the CARDEA series. It is capable of maximum sequential read speeds of over 13,000 MB/s and write speeds exceeding 12,000 MB/s, and together with a maximum storage capacity of 4 TB, it will be the highest performance PCIe Gen5 flagship SSD on the market when launched. The T-FORCE PCIe Gen5 SSD will support the latest NVMe 2.0 and will be enhanced using proprietary technology to improve durability and product lifespan. The new hardware and support will enable its Gen5 SSDs to fully meet the needs of the gaming computing market, video streaming, augmented reality (AR), virtual reality (VR), and other next-gen digital domains. TEAMGROUP plans to mass-produce T-FORCE CARDEA series' PCIe Gen5 SSD in 2022 Q3, and will also launch a variety of efficient cooling solutions for the PCIe Gen5 SSD. Together with its global motherboard manufacturing partners, TEAMGROUP will continue to refine specifications and technologies of the latest generation of SSDs to meet market demands and provide consumers around the world with a stable and rapid storage experience.

At the launch of the Z690 chipset, ASRock didn't reveal its flagship Aqua model, but now the company has finally announced not just one, but two Z690 Aqua models, one "regular" SKU and one OC SKU. The main difference between the two is that the OC SKU only has two DIMM slots, since as we know, DDR5 is extra finicky when it comes to running at high clock speeds and two DIMM boards allow for higher DRAM clock speeds, This also means that all the major board makers have now launched a two DIMM product geared towards overclockers.

The two boards are mostly identical in terms of features otherwise, but there are some minor differences that we'll point out later. As expected, both boards feature a large, nicely integrated liquid cooling plate for the CPU and the 20 phase 105 A SPS VRM setup. ASRock has gone for a 12 layer PCB, not something you see in a lot of consumer motherboards and it's unclear if it'll bring any real benefits, but as this is a premium motherboard, it's a nice touch. The heatsink of the top-most M.2 drive is home to a small OLED display, but ASRock hasn't listed the size on its website.

congatec - a leading vendor of embedded and edge computing technology - introduces the 12th Generation Intel Core mobile and desktop processors (formerly code named Alder Lake) on 10 new COM-HPC and COM Express Computer-on-Modules. Featuring the latest high performance cores from Intel, the new modules in COM-HPC Size A and C as well as COM Express Type 6 form factors offer major performance gains and improvements for the world of embedded and edge computing systems. Most impressive is the fact that engineers can now leverage Intel's innovative performance hybrid architecture. Offering of up to 14 cores/20 threads on BGA and 16 cores/24 threads on desktop variants (LGA mounted), 12th Gen Intel Core processors provide a quantum leap [1] in multitasking and scalability levels. Next-gen IoT and edge applications benefit from up to 6 or 8 (BGA/LGA) optimized Performance-cores (P-cores) plus up to 8 low power Efficient-cores (E-cores) and DDR5 memory support to accelerate multithreaded applications and execute background tasks more efficiently.

BIOSTAR, a leading manufacturer of motherboards, graphics cards, and storage devices today, introduces the brand new B660 series motherboards geared to run the latest Intel 12th generation processors. Available in 3 distinct models, B660GTA, B660GTQ, and the B660GTN, BIOSTAR's new B660 series motherboards carry the legacy of their predecessors as the best performance motherboards available in the market. The B660 Series motherboards are the best mid-range motherboard series in the market for business users, content creators and casual gamers alike that works the best with Intel 12th generation processors.

Featuring PCIe 5.0, the latest leap in data transmission technology, and PCIe M.2 4.0 storage technology, BIOSTAR's B660 series motherboards are robust and functional, designed to suit any use case scenario. The B660GTA, the B660GTQ, and the B660GTN motherboards come in 3 distinct form factors ATX, uATX, and Mini-ITX respectively. For users to select between their favorite build sizes.

ASUS has today announced the ROG Loki SFX-L power supply lineup during their ROG CES 2022 livestream with models ranging from 750 W to 1200 W with 80 Plus Platinum ratings. These new power supplies are designed for small-form-factor gaming builds featuring high-core processors and power-hungry graphics cards. The units include a 16-pin power cable for PCIe 5.0 GPU support and feature a 120 mm dual-ball bearing axial-tech fan with PWM control to manage noise and keep temperatures low. The Loki power supplies will feature integrated ASUS Aura ARGB lighting with Aura Sync compatibility and custom ROG heatsinks cover critical components. The ASUS ROG Loki SFX-L power supply lineup will be available to purchase worldwide in Q1/Q2 2022.

XPG, a fast-growing provider of systems, components, and peripherals for Gamers, Esports Pros, and Tech Enthusiasts, today announces it will be on location in Las Vegas during the 2022 CES show (Jan. 5-8). Under the theme of 'Dawn of a New Xtreme,' it will be showing off the latest products from XPG and ADATA including gaming systems, peripherals, accessories, and components, such as the newest DDR5 memory, PCIe 5.0 solid state drives (SSD) and the latest state-of-the-art Platinum PSU.

XPG's and ADATA are gearing up to bring next-generation performance to gamers and other users with its latest PCIe Gen5x 4 SSDs and DDR5 DRAM. At the booth, visitors will get a first look at the XPG LANCER RGB DDR5 and CASTER RGB DDR5 that deliver next-level performance of up to 6000 MT/s and 7000 MT/s, respectively. ADATA's industrial-grade DDR5-4800 U/SO/R-DIMM memory with on-die ECC & PMIC will also be on display.

Samsung Electronics Co., Ltd., the world leader in advanced memory technology, today announced that it has developed the PM1743 SSD for enterprise servers, integrating the PCIe (Peripheral Component Interconnect Express) 5.0 interface with Samsung's advanced sixth-generation V-NAND.

"For over a decade, Samsung has been delivering SATA, SAS and PCIe-based SSDs that have been recognized for outstanding performance and reliability by leading enterprise server customers including corporations, governments and financial institutions," said Yong Ho Song, Executive Vice President and Head of the Memory Controller Development Team at Samsung Electronics. "The introduction of our PCIe 5.0 SSD, along with PCIe 6.0-based product developments that are underway, will further solidify our technological leadership in the enterprise server market."

There has been a lot of discourse about the cost of Z690 motherboards, so we decided to ask around to find out what has changed compared to the previous generation of Intel motherboards. There are obviously several factors that have come together to create something like the perfect storm, as the simple answer is that all parts combined have resulted in more expensive motherboards, but there are a few key components that are major contributing factors.

A lot of speculation has been about the cost of the PCB itself and although it's correct that the PCB is a contributing factor, we're talking about a couple of dollars in extra cost, not only for the higher quality PCB materials themselves but also the fact that boards with DDR5 memory are more costly to produce, as more care needs to be taken with the overall design. However, the big surprise to us is that the single part that has driven up the cost the most is the physical LGA-17xx CPU socket and retention mechanism, which is apparently around four times as expensive as the LGA-12xx socket.

Based on the Intel 600-series chipset leak earlier today, it was easy to conclude that all 600-series motherboards would be getting PCIe 5.0 support, but alas, that is not the case. We've already seen some rock bottom Z690 without PCIe 5.0 and thanks to Videocardz we now know that Gigabyte's B660 Gaming X DDR4 will also lack PCIe 5.0 support.

How do we know this board doesn't support PCIe 5.0? First of all, the x16 slot closest to the CPU uses one of Gigabyte's older reinforced PCIe 4.0 slots, whereas all of its Z690 boards with PCIe 5.0 support uses a new, white type of slot. These slots use a visually different kind of reinforcement as well, even if it's only marginally different. However, Gigabyte doesn't appear to be silk screening PCIe 5.0 on supported boards and it's possible that they're using a different slot vendor for the B660 boards.

The ASUS PRIME B660-PLUS D4 has allegedly been inadvertently sent instead of an ASUS Z690 motherboard to a reviewer for Alder Lake testing. The reviewer provided images of the product packaging to VideoCardz which clearly show a label indicating PCIe 4.0 support. This has come as a surprise as the Intel 12th Generation Alder Lake processors include attached PCIe 5.0 lanes separate from the chipset. The B660 chipset will target the mid-range market so the lack of PCIe 5.0 support on this specific motherboard may be a cost-cutting or artificial separation measure.

The first consumer PCIe 5.0 graphics cards and SSDs aren't expected to arrive until H1 2022 and will likely come at a significant premium so the exclusion of support could make sense for a more budget-oriented platform. Intel is also preparing to launch a higher-end H670 chipset which is expected to include PCIe 5.0 support. We cannot be sure if this label is accurate and if the lack of PCIe 5.0 support will apply to all B660 motherboards so take these rumors with a healthy dose of skepticism until the chipset and motherboards are officially unveiled.

Although we did mention the MSI MEG Z690 GODLIKE in our Z690 motherboard roundup piece, until now, we didn't really know what the board looked like, since MSI has been very shy when it comes to sharing details of the board. However, now the company has released a couple of pictures of the board, courtesy of the board winning a pair of CES 2022 Innovation Honoree awards.

We're still short on details of this monster board, as it'll measure 305x310 mm, which almost puts it in SSI EEB territory. The board has what can only be described as a full cover, since outside of the CPU socket and some of the components near it, you really can't see anything. On top of the memory modules, MSI has gone for a display of some kind that seems to display some system information, such as the CPU type and current clock speed. The location of the display doesn't seem to be ideal though, but it's said to measure in at 3.5-inches.

A new era for flash used in servers and storage is here, and it's being introduced by KIOXIA America, Inc. Today, the company announced the industry's first lineup of Enterprise and Datacenter Standard Form Factor (EDSFF) E3.S SSDs designed with PCIe 5.0 technology - the CD7 Series. Building on the KIOXIA E3.S full-function development vehicle that received a 'Best in Show' award at last year's Flash Memory Summit, the CD7 Series E3.S increases flash storage density per drive for optimized power efficiency and rack consolidation.

Breaking free from the design limitations of the 2.5-inch form factor, the EDSFF E3 family is optimized for the needs of high-performance, highly efficient servers and storage. EDSFF enables the next generation of SSDs to address future data center architectures, while supporting a variety of new devices and applications. It provides improved airflow and thermals, signal integrity benefits, and options for larger SSD capacity points. EDSFF standardizes status LEDs on the drive, which eliminates the need for LEDs on the drive carriers. Support for higher E3.S power budgets than 2.5-inch form factor SSDs and better signal integrity allows EDSFF to deliver the performance promised by PCIe 5.0 technology and beyond.

GIGABYTE Technology, one of the top global manufacturers of motherboards, graphics cards, and hardware solutions, announces today Z690 AORUS TACHYON motherboard with the latest 12th generation Intel Core i9 12900K processor and LN2 liquid nitrogen heat dissipation set the world-leading overclocking records of dual 8 GHz, which is 8000 MHz on i9-12900K CPU and 8300 MHz on DDR5 memories. GIGABYTE Z690 AORUS TACHYON motherboard designed by world-famous overclocker equips with plenty functions exclusively for extreme overclockers, which promise overclockers higher scores with easier operation, establishing GIGABYTE's leading role in overclocking on the Z690 motherboards.

"GIGABYTE demonstrate its determination of returning to the overclocking glory with the launch of the Z590 AORUS TACHYON motherboard, and the release of the Z690 AORUS TACHYON further verified the continuation of GIGABYTE's overclocking pedigree on motherboards." said Jackson Hsu, Director of the GIGABYTE Channel Solutions Product Development Division. "Z690 AORUS TACHYON proves GIGABYTE's strong R&D strength on LN2 overclocking by the world records it created. We will definitely use Z690 AORUS TACHYON to create better overclocking performance, and can't wait to see that overclockers around the world use this motherboard to break more world records.

ASUS' upcoming Z690 motherboards have been revealed and the company has put its product pages as well as posted additional details of most of its upcoming boards, which gives us a good look at where they're heading with the Z690-series. As mentioned before, ASUS has something like 30 different SKUs, so expect to spend some time to go over all the different models.

We're not going to spend too much time on the details right now and we suggest you head over to the source links and read up on the nitty gritty bits there. A few things to point out though is that ASUS appears to have worked on its own memory overclocking standard called AEMP, which is said to handle memory that lacks XMP profiles, which implies that Intel might not have had time to work with memory manufacturers to get XMP ready for DDR5. ASUS' ROG Maximus and ProArt motherboards feature dual PCIe 5.0 slots, whereas all other models only feature one, albeit obviously a single x16 or dual x8 slots. Some boards will feature Thunderbolt 4 and ASUS has even put a front connector on some of those, although we're not sure how that will work in terms of case compatibility.

With only about two weeks until Intel's official launch of its Alder Lake CPUs and the Z690 chipset, it would appear that the leaks are getting kicked up a notch and the latest leak is a rather intriguing one to say the least. Gigabytes Z690 Aorus Xtreme is a rather unusual looking motherboard, in as much as you can make a motherboard that look unusual, least not with regards to the slot placement, but also for the peculiar cover over the DRAM slots.

It's not quite clear what the function of the memory cover is beyond aesthetics, if any, but it seems unlikely that it has an integrated LCD screen, although not unthinkable. Pretty much the entire board is covered in either metal or plastic, which makes this look like a real chore to build, at least if you're planning on using M.2 SSDs. The standout feature here is that the board features a pair of PCIe 5.0 x16 slots, although as per current consumer boards, the bandwidth would be shared between the slots if the second one is used, so both slots end up as x8 slots. There's also a third PCIe x16 slot, but no indication of the bandwidth available to it.

The first mainstream Z690 boards have made an appearance online in the shape of ASUS' Prime Z690-P series. As with most of ASUS' Prime boards, these are fairly basic boards, but shouldn't leave the average user wanting for much, although ASUS has made a rather odd layout decision on these boards that we can't quite grasp.

The four models on display are the Prime Z690-P WiFi D4, Prime Z690-P D4, Prime Z690-P and Prime Z690M-Plus D4. The last model doesn't strictly belong to the Prime Z690-P family, but it follows a pretty similar overall design. The Prime Z690-P is obviously the only DDR5 SKU among the four boards, although from what we can tell, there are no other differences between it and the equivalent DDR4 model.

Marvell today announced the expansion of its industry-leading 5 nm data infrastructure platform with the launch of the industry's first 5 nm 50G PAM4 device for the carrier market, the Prestera DX 7321 Ethernet switch. The new switch builds on the success of the Prestera carrier-optimized portfolio and is ideal for 5G fronthaul and edge connectivity. In concert with this, Marvell's 5 nm OCTEON 10 DPU family, incorporating industry-leading hardware accelerators, is now sampling. By utilizing the industry's leading advanced process geometry, the Marvell Prestera switch and OCTEON DPU deliver 50% lower power than existing offerings, enabling new infrastructure solutions for next-generation carrier edge networks and RAN deployment models.

With the addition of the 5 nm Prestera device, the expanded carrier-optimized switch portfolio now comprises four cutting-edge Ethernet switches that scale port speeds from 1 Gbps to 400 Gbps with aggregate bandwidths ranging from 200 Gbps to 1.6 Tbps. The newest offering enhances Marvell's 5G solutions for Open RAN, vRAN and traditional RAN architectures, with Class D precision time protocol (PTP), which provides more timing headroom to enable larger cell coverage radius. The switch device incorporates integrated MACsec security and advanced telemetry to facilitate network visibility and automation.