Today, thanks to Yuuki_Ans on the Chinese Bilibili forum, we have more information about the upcoming "Avenue City" platform that powers Granite Rapids and Sierra Forest. Intel's forthcoming Granite Rapids and Sierra Forest Xeon processors will diverge the Xeon family into two offerings: one optimized for performance/core equipped with P-cores and the other for power/core equipped with E-cores. The reference platform Intel designs and shares with OEMs internally is a 16.7" x 20" board with 20 PCB layers, made as a dual-socket solution. Featuring two massive LGA-7529 sockets, the reference design shows the basic layout for a server powered by these new Xeons.

Capable of powering Granite Rapids / Sierra Forest-AP processors of up to 500 Watts, the platform also accommodates next-generation I/O. Featuring 24 DDR5 DIMMs with support for 12-channel memory, with memory speeds of up to 6400 MT/s. The PCIe selection includes six PCIe Gen 5 x16 links supporting CXL cache coherent protocol and 6x24 UPI links. Additionally, we have another piece of information that Granite Rapids will come with up to 128 cores and 256 threads in both regular and HBM-powered Xeon Max flavoring. You can see storage and reference platform configuration details on the slides below.

Intel has quietly canceled its Thunder Bay hybrid system-on-chip (SoC) that combines standard general-purpose CPU Cores and Movidius Vision Processing Unit (VPU) cores. Such chips were aimed at commercial and Internet-of-Things (IoT) applications that relied on computer vision acceleration and edge-computing applications. According to the latest report, it appears that Thunder Bay is canceled as Intel has released a set of patches that removed the Thunder Bay code from the Linux kernel.

Intel kept Thunder bay details well hidden and while earlier rumors pointed to a combination of Intel Xeon x86 CPU cores and Movidius VPU cores, the only Thunder Bay support in Linux patches where showing a combination of ARM Cortex-A53 low-power cores with the Movidius VPU. Intel acquired Movidius back in 2016 and while it is not something that Intel talks about often, there are several products based on Movidius VPUs, including Neural Compute Stick, Intel drones, the Intel RealSense Tracking Camera, and most recently, the Intel Movidius 3700VC VPU, formerly Keem Bay. In any case, it appears that Intel is abandoning an idea of combining general purpose x86 CPU cores and Movidius VPU cores.

EUROCOM Raptor X17 Mobile Supercomputer is powered by Intel i9-13900HX 24C/32T Processor and NVIDIA GeForce RTX 4090 16 GB DDR6 GPU. It comes with 10 I/O ports and supports up to 24 TB of SSD storage in RAID 0/1/5, 2x Thunderbolt 4 and a stunning choice of 17.3-inch UHD, QHD and FHD LCD panels ranging up to 300 Hz. Perfect for 3D Modeling and Rendering applications.

The Raptor X17 is the newest, mobile supercomputer class laptop from Eurocom. Powered by Intel HM770 chipset, this mobile supercomputer weighs 3.2 kg, making it much easier to move around for on-the-go deployment.

Intel Graphics today released the latest version of Arc GPU Graphics Drivers. Version 101.4148 beta comes with Game On (day-zero) optimization for Deceive Inc., and Diablo 4 (beta); and optimization for Sons of The Forest. It fixes an issue where Riftbreaker (DX12) Xbox Game Pass version exhibits a rare black line display corruption; and lower than expected performance for Resident Evil 4 Remake Chainsaw Demo (DX12). The company also took the opportunity to highlight that with recent price cuts for the Arc A750 down to $249, it offers anywhere between 54% to 76% higher performance-per-Dollar than the competing GeForce RTX 3060.

DOWNLOAD: Intel Arc GPU Graphics Drivers 101.4148 beta

GIGABYTE AORUS invites gamers and fans to attend AORUSVERSE, a vast gaming universe packed with the latest AORUS gaming hardware and gears, at PAX EAST 2023. The triumphant return to Boston will also feature fun activities and exciting esports challenges throughout the booth, where visitors can get their hands on the latest tech and participate in various events to win prizes.

At AORUSVERSE, attendees can explore the 2023 gaming laptop lineup, featuring the flagship AORUS 17X and 15X. These laptops are powered by the latest Intel 13th gen CPU and NVIDIA GeForce RTX 40 laptop GPU, delivering a giant leap in gaming performance. They also come with QHD displays with up to 240 Hz refresh rates for a smooth and fluid gameplay. For those looking for a balance of performance and portability, the non-X variants, AORUS 17 and 15, will also be available onsite for attendees to test personally.

Intel's latest Xeon processors based on Sapphire Rapids uArch have arrived in the hands of overclockers. Last week, we reported that the Intel Xeon W9-3495X is officially a world record holder for achieving the best scores in Cinebench R23 and R20, Y-Cruncher, 3DMark CPU test, and Geekbench 3. However, today we have another extreme overclocking attempt to beat the world record, with little more details about power consumption and what the new SKU is capable of. Elmor, an overclocker working with ASUS, has tried to break the world record and overclocked the Intel Xeon W9-3495X CPU to 5.5 GHz on all 56 cores. What is more impressive is the power that the processor can consume.

With a system powered by two Superflower Leadex 1,600 Watt power supply units, the CPU consumed almost 1,900 Watts of power from the wall. To manage to cool this heat output, liquid nitrogen was used, and the CPU stayed at a cool negative 95 degrees Celsius. The motherboard of choice for this attempt was ASUS Pro WS W790E-SAGE SE, paired with eight GSKILL Zeta R5 DDR5 R-DIMMs modules. And results were incredible, as the CPU achieved 132,220 points in Cinebench R23. However, the world record of the previous week has remained intact, as Elmor 's result is a bit behind last week's score of 132,484 points. Check the video below for more info.

In the release of Python 2.0, contributions from Intel using Intel Extension for PyTorch, oneAPI Deep Neural Network Library (oneDNN) and additional support for Intel CPUs enable developers to optimize inference and training performance for artificial intelligence (AI).

As part of the PyTorch 2.0 compilation stack, the TorchInductor CPU backend optimization by Intel Extension for PyTorch and PyTorch ATen CPU achieved up to 1.7 times faster FP32 inference performance when benchmarked with TorchBench, HuggingFace and timm. This update brings notable performance improvements to graph compilation over the PyTorch eager mode.

ASUS today announced that ASUS W790 workstation motherboards support ECC R-DIMM DDR5 RAM at up to 6800 MHz, unleashing class-leading overclocking capability to unlock even more performance potential. ASUS W790 workstation motherboards feature the memory capability of up to 2 TB RDIMM DDR5 memory to power the next generation of compute-intensive professional workloads. In addition, ASUS have been working with market-leading memory partners including G.SKILL, Kingston, and V-Color to establish multiple memory validations and to widely support 8-channel and quad-channel memory architectures. The Pro WS W790E-SAGE SE now supports up to 128 GB (8x16 GB) at 6800 MT/s, and the Pro WS W790-ACE supports up to 64 GB (4x16 GB) of DDR5 RAM at 6800 MT/s. This painstakingly cultivated collection of device validations helps ensure that an ASUS W790 workstation build supports a wide variety of scenarios.

Choosing the right CPU cooler for a workstation build can also be a challenge, because CPU sockets and form factors are different from those used in desktops. Therefore, ASUS has partnered with cooling experts including Enermax, EK and Noctua to offer air and liquid coolers for the LGA4677 socket used by the latest Intel Xeon workstation processors. These solutions are designed to offer optimal heat transfer and heat dissipation capacity through the cooler base and then to coolant tubes, heat pipes and radiator fins.

DFI, the global provider of industrial computers, has introduced its ECX700-AL ruggedized x86 PC for outdoor edge computing applications. Its impeccable performance capabilities, compact size, and high durability make it the ideal solution for outdoor operations. The ECX700-AL is a rugged computer with a waterproof and dust-proof design to withstand harsh environments and any type of climate. Stable performance is ensured with its wide operating temperature of -40 to 70 degrees Celsius. It is also equipped with a smart vent that drains water automatically, preventing typical problems such as condensation within the system or excessive heat.

An external SIM slot provides convenience for users when changing SIM cards, and the high-gain antenna delivers a high-quality signal. Versatile I/O settings including a combo port that supports 2 CANs and 2 COMs as well as selective display ports (HDMI/VGA) help integrators meet demands for various applications.

Supermicro, Inc., a Total IT Solution Provider for Cloud, AI/ML, Storage, and 5G/Edge, is announcing the latest addition to its revolutionary ultra-high performance, high-density petascale class all-flash NVMe server family. Supermicro systems in this high-performance storage product family will support the next-generation EDSFF form factor, including the E3.S and E1.S devices, in form factors that accommodate 16- and 32 high-performance PCIe Gen 5 NVMe drive bays.

The initial offering of the updated product line will support up to one-half of a petabyte of storage space in a 1U 16 bay rackmount system, followed by a full petabyte of storage space in a 2U 32 bay rackmount system for both Intel and AMD PCIe Gen 5 platforms. All of the Supermicro systems that support either the E1.S or E3.s form factors enable customers to realize the benefits in various application-optimized servers.

Kontron, a leading global provider of IoT/Embedded Computer Technology (ECT), will present an Mini-ITX motherboard based on Intel Core i3 processors as well as Intel N-series processors with latest UHD Gen12 graphics at embedded world 2023, thus expanding its product family of compact, power-saving motherboards "Designed and Made in Germany." The K3931-N Mini-ITX is available in four variants with different CPUs and is particularly suitable for use in the areas of casino gaming, digital signage, kiosk, POS/POI, ticketing as well as for applications in industrial and medical technology.

The Intel Core i3 and Intel N series processors set new standards for x86 processors in the power range from 6 W to 15 W. They use the same Efficient-cores and Intel UHD graphics as the 12th Gen Intel Core processors based on the Xe architecture. This makes it easier to port applications and solutions across Intel CPU performance and power ranges. Support for Intel AVX2 and Intel Deep Learning Boost - combined with up to 32 execution units - provides accelerated deep learning inference and media processing for diverse edge applications.

ASRock, with the help of legendary overclocker Splave have used their new W790 WS to take Intel's Sapphire Rapids based Xeon w7-2495X as high as 5.3 GHz to grab a pile of first place global performance scores on HWBOT. In total they've secured 14 global records in the 24-core category, and 18 more records in the w7-2495X hardware category. What is perhaps slightly more impressive is that they managed to exceed the officially supported DDR5-4800 by 2000 MT/s, achieving a frequency of 3400 MHz (6800 MT/s) with a CAS Latency of 32. The results were all produced with a custom water cooled configuration, with many benchmarks needing below 5 GHz clocks to remain stable enough for validation, so there is certainly room for higher scores under more exotic cooling solutions.

One of the more interesting details presented in the screenshots of certain benchmark results such as Y-Cruncher and Cinebench R15 is the gargantuan power draw exhibited by the w7-2495X. With a rated Vcore of 1.22v the 24-core sucks down over 500 W. We have already seen the 56-core w9-3495X pull a full kilowatt in early overclocking sessions, so the power draw here is not entirely without merit. Needless to say while Intel is finally offering a compelling HEDT lineup of unlocked processors, you may not necessarily be able to squeeze much out of them before tripping your breakers. As for ASRock's records it's still early days for the new Xeon W lineup and there will be a revolving door of world record holders before the final ounce of performance is squeezed from these chips.

On Jan. 10, 2023, Intel officially launched its 4th Gen Intel Xeon Scalable processors (code-named Sapphire Rapids) for data center customers around the globe. Fueled by years of hard work and focus, Intel delivered its highest quality Xeon product ever, and is already seeing a rapid ramp and the greatest-ever level of customer and partner support.

Launch day brought to market a new Intel leadership product bolstered by significant advancements in performance with industry-leading accelerators, increased core count and improved performance per watt. With a strong new product in the market, customer adoption of these new technologies has been swift. In just eight weeks since launch, Intel has executed the most ever design wins for any Xeon family and the most ever platforms available and shipping in this short time-to-market window.

ASUS IoT, the global AIoT solution provider, today announced PE6000G - the industry's first edge AI system that supports up to a 450 W GPU for high-throughput, demanding GPU-computing applications. Powered by a 12th Gen Intel Core processor and paired with the Intel R680E chipset, PE6000G supports up to 450 W graphics card power and is ready for the latest PCI Express (PCIe ) 5.0 standard, meeting requirements for enabling real-time AI inference at the edge.

Demand for high-performance GPU-accelerated solutions at the edge is increasing to satisfy mission-critical AI applications. Users who require rugged and durable systems to withstand shocks and vibration, plus wide-range operating temperatures to accommodate high GPU computing power, face the dilemma of finding solutions that provide the necessary specifications to meet physical space constraints, along with heat dissipation and power consumption within a rugged mechanical design. The all-new PE6000G series strikes a balance between all these needs for deployment in a variety of challenging environments.

Microsoft Azure announced their new ND H100 v5 virtual machine which packs Intel's Sapphire Rapids Xeon Scalable processors with NVIDIA's Hopper H100 GPUs, as well as NVIDIA's Quantum-2 CX7 interconnect. Inside each physical machine sits eight H100s—presumably the SXM5 variant packing a whopping 132 SMs and 528 4th generation tensor cores—interconnected by NVLink 4.0 which ties them all together with 3.6 TB/s bisectional bandwidth. Outside each local machine is a network of thousands more H100s connected together with 400 GB/s Quantum-2 CX7 InfiniBand, which Microsoft says allows 3.2 Tb/s per VM for on-demand scaling to accelerate the largest AI training workloads.

Generative AI solutions like ChatGPT have accelerated demand for multi-ExaOP cloud services that can handle the large training sets and utilize the latest development tools. Azure's new ND H100 v5 VMs offer that capability to organizations of any size, whether you're a smaller startup or a larger company looking to implement large-scale AI training deployments. While Microsoft is not making any direct claims for performance, NVIDIA has advertised H100 as running up to 30x faster than the preceding Ampere architecture that is currently offered with the ND A100 v4 VMs.

ASUS today announced the ExpertCenter PN64-E1, a mini PC powered by up to a 13th Gen Intel Core processor with up to DDR5-4800 memory and Intel Iris Xe Graphics. The PN64-E1 features two PCIe 4.0 x4 M.2 slots, five USB ports, two Thunderbolt 4 ports with DisplayPort 1.4 support, two HDMI, plus a configurable port for flexibility. In addition, the chassis provides quick access to the storage and memory modules for easy upgrades. Compared to the previous generation, the 13th Gen Intel Core processor that powers the PN64-E1 delivers up to 9% faster performance to enable smooth and efficient processing of intensive business applications. DDR5 memory offers considerably faster data speeds over DDR4, ensuring efficient productivity.

Apex Storage, a new company in the storage world, has announced that its X21 add-in-card (AIC) has room for 21 (you read that right) PCIe 4.0 M.2 NVMe SSDs. The card supports up to 168 TBs with 8 TB M.2 NVMe SSDs and 336 TBs of storage with future 16 TB M.2 SSDs drives and can withstand speeds of up to 30.5 GB/s. Packed inside a single-slot, full-length, full-height AIC, the X21 card is built for a snug fit inside workstations and applications such as machine learning and hyper-converged infrastructure that enterprises need to develop inside servers and workstations across the site.

The X21 AIC has 100 PCIe lanes on the board, which indicates the presence of a PCIe switch, likely placed under the heatsink. To power all the storage, the PCIe slot itself needs to be more, and the card also has two 6-pin PCIe power connectors that provide 225 Watts of power in total. Interestingly, the heatsink is passively cooled, but Apex Storage suggests that there should be an active airflow with a minimum of 400 LFM to ensure the regular operation of the card. In the example application, the company laid out X21 with Samsung's 990 Pro SSDs; however, the card also supports Intel Optane drives. Read and Write IOPS are higher than 10 million. Additionally, the average read and write access latencies are 79 ms and 52 ms. Apex Storage didn't reveal the pricing and availability of the card; however, expect it to come with a premium.

Intel is preparing a revision to the design of the 12+4 pin ATX 12VHPWR power connector, specifically its male connector, to improve connector mating and retention; and to minimize a potential fire hazard asiring from improper retention or mating. The male connector design revision will retain full compatibility with female connectors of devices in circulation. It essentially sees the pins of the male connector switch from a "three dimple" joint to a "push-spring" type joint. This design change makes the male connector much more resilient to weakening contact from pull forces arising from cable bending. The resulting connector would offer superior structural integrity of the connection.

Intel in February launched a couple of China-exclusive 13th Gen Core SKUs to highly optimize their price-performance for the PC gaming crowd there. These included the Core i5-13490F (6P+4E but with 24 MB L3 cache), and Core i7-13790F (8P+8E but with 33 MB L3 cache). In the wake of increased competition from AMD with its non-X Ryzen 5 7600, Ryzen 7 7700, and price-adjustments for its 7600X and 7700X, Intel is giving these SKUs slight price cuts. The i5-13490F launched at ¥1,599 (roughly $230), but is now retailing for ¥1,499 (roughly $200). There's no word on if these SKUs will launch outside China. The idea behind them is to increase the L3 cache sizes without touching the core-counts or clock-speeds, to improve cache-sensitive gaming performance. These chips are packaged in special "dark" retail boxes as shown below.

Over the past couple days ASRock has been busy slashing prices on their lineup of Arc graphics cards. Currently (on Newegg at least) you can now find their Challenger Arc A380 for $120, down from $150, the Challenger Arc A750 8 GB at $240, down from $290, and the flagship Phantom Gaming Arc A770 8 GB at $270, down from $320. The new pricing may be in response to prior price drops by Intel, and possibly hints at more to come.

With its new price ASRock's Arc A770 8 GB in particular is placed in a more advantageous position against NVIDIA's RTX 3060 which has only recently seen reduced prices and still lists for well above $300, even $400 for some partner models. AMD is still placed quite competitively thanks to ASRock themselves listing their RX 6600 XT offerings for $275, making the choice between the RX 6600 XT and Arc A770 8 GB a more difficult one. The RX 6650 XT could also be considered as it regularly appears for $290, a mere $20 more than the revised A770 8 GB pricing, and provides a measurable (if minimal) uplift in performance.

When Intel announced the appearance of the 4th generation Xeon-W processors, the company announced that the clock multiplier was left unlocked, available for overclockers to try and push these chips even harder. However, it was only a matter of time before we saw the top-end Xeon-W SKU take a chance at beating the world record in Cinebench R23. The Intel Xeon W9-3495X SKU is officially the world record score holder with 132,484 points in Cinebench R23. The overclocker OGS from Greece managed to push all 56 cores and 112 threads of the CPU to 5.4 GHz clock frequency using liquid nitrogen (LN2) cooling setup. Using ASUS Pro WS W790E-SAGE SE motherboard and G-SKILL Zeta R5 RAM kit, the OC record was set on March 8th.

The previous record holder of this position was AMD with its Threadripper Pro 5995WX with 64 cores and 128 threads clocked at 5.4 GHz. Not only did Xeon W9-3495X set the Cinebench R23 record, but the SKU also placed the newest record for Cinebench R20, Y-Cruncher, 3DMark CPU test, and Geekbench 3 as well.

BIOSTAR, a leading manufacturer of motherboards, graphics cards, and storage devices today, is thrilled to showcase the latest B760MZ-E PRO motherboard. The BIOSTAR B760MZ-E PRO motherboard is a versatile solution that meets diverse pro-user demands. Whether you are a professional creator seeking the hottest mid-range hardware on the market, a business professional requiring reliable and efficient computing power, or a casual user needing a seamless computing experience, the B760MZ-E PRO has you covered. Its cutting-edge technology, advanced DDR5 memory support, lightning-fast transfer speeds, and connectivity options make it the ultimate choice for those seeking a high-performance motherboard that can handle anything thrown its way.

Built on Intel's B760 single-chip architecture, The BIOSTAR B760MZ-E PRO motherboard supports the latest 12th/ 13th Generation Intel Core Processors, providing robust and stable performance for demanding applications. Its 4-DIMM DDR5 memory support and PCIe M.2 4.0 (64 Gb/s) lightning-fast transfer speeds offer unrivaled computing and file transfer efficiency. Ideal for content designers, business professionals, and casual users seeking the best-performance motherboard of this range.

MSI is a top player in the ICT industry. With over 35 years of experience in R&D. By introducing AI systems, IoT, automation, and remote operations. Leveraging its many years of experience in ODM/OEM and its production capabilities, MSI delivers products that apply to smart transportation, smart cities, innovative automation, digital signage and intelligent retailing, offering top-quality, integrated solutions that best suit client needs. MSI commercial monitors offer crisp visuals and a range of features to enhance the computing experience.

MS-98M3 is the embedded 3.5" SBC that supports multiplexing with 11th generation Intel IOTG Core processor, codenamed Tiger Lake-UP3 series of ultra-low power and high-performance, providing reliability and high durability with lower power consumption, and actively deploy in multiple applications such as AMR-AI-PJ-UVGI Robot, which was verified by Texcell/France to effectively inactivate the new SARS-COV2 virus within seconds.

Today, Intel announced that the LF AI & Data Foundation Technical Advisory Council accepted Open Federated Learning (OpenFL) as an incubation project to further drive collaboration, standardization and interoperability. OpenFL is an open source framework for a type of distributed AI referred to as federated learning (FL) that incorporates privacy-preserving features called confidential computing. It was developed and hosted by Intel to help data scientists address the challenge of maintaining data privacy while bringing together insights from many disparate, confidential or regulated data sets.

"We are thrilled to welcome OpenFL to the LF AI & Data Foundation. This project's innovative approach to enabling organizations to collaboratively train machine learning models across multiple devices or data centers without the need to share raw data aligns perfectly with our mission to accelerate the growth and adoption of open source AI and data technologies. We look forward to collaborating with the talented individuals behind this project and helping to drive its success," said Dr. Ibrahim Haddad, executive director, LF AI & Data Foundation.

After GUNNIR showed the same product back in January, ADLINK is now offering both Intel Arc A-series GPUs in MXM form factor. The MXM (Mobile PCI Express Module) is a standardized form factor that is used mostly in laptops and some small form factor PCs. Product pages confirm that ADLINK offers both the Intel Arc A370M and the Intel Arc A350M in MXM form factor.

According to specifications The ADLINK MXM-AXe, as the product is called, is MXM 3.1 Type A based on Intel Arc GPU, packing 8 Xe-cores, 128 Execution Units, 4 GB of GDDR6 memory, and TDP of 35-50 W, which is pretty much standard for the Arc A370M GPU. The company also offers the same product with A350M GPU with TDP of 25-35 W. With decent power efficiency, full AV1 hardware encoding, and support for up to four 4K displays, such a GPU would be perfect for small form factor machines, and could be even a decent upgrade for some laptops.