Intel is still in the work of testing and certification for their more complex graphics products based on their Xe microarchitectures, and that means that some tests are being done in well-known benchmarking platforms. Case in point, an Intel Xe GPU with a reported 128 EUs (1024 shading units), 3 GB of memory, and a clockspeed of 1.4 GHz ran through Geekbench, where it scored an... interesting 9,311 points in the OpenCL test.

This is more likely than not an engineering sample, considering that Intel's Iris Xe MAX wrapped in its Tiger Lake package can score up to 23,000 points. It is currently unclear if this particular Xe manifestation is running on Intel's Xe-LP or Xe-HPG architecture. This might be Intel's DG-2 product, which offers higher performance than their DG-1 discrete graphics card that is only available for system integrators.

QNAP Systems, Inc., a leading computing, networking and storage solution innovator, today announced the desktop smart edge PoE Switch - QGD-3014-16PT. With sixteen 30-watt Gigabit PoE ports, two 2.5GbE host management ports, Intel Celeron J4125 quad-core 2.0 GHz processor, and four 3.5-inch SATA drive bays, the QGD-3014-16PT supports QVR Pro, HBS 3 and QuWAN SD-WAN to integrate surveillance deployment, video storage computing, and multi-site remote management to offer SMBs innovative intelligent IP surveillance infrastructure and remote backup solutions.
"Expanding multi-site surveillance networks can be costly and involve large amounts of equipment with low transmission efficiency between multiple remote devices - not to mention the complexities involved in deployment and management. " said Daniel Hsieh, QNAP Product Manager, adding, "the QGD-3014-16PT desktop Smart Edge PoE Switch integrates PoE, surveillance capabilities, and data backup management to simplify these requirements and increase the transmission and backup efficiency of surveillance videos."

The percentage distribution of 2021 NAND Flash bit demand by application currently shows that client SSD accounts for 31%, enterprise SSD 20%, eMMC/UFS 41%, and NAND wafer 8%, according to TrendForce's latest investigations. TrendForce expects NAND Flash ASP to undergo QoQ declines throughout 2021, since the number of NAND suppliers far exceeds DRAM suppliers, and the bit supply remains high. As Samsung, YMTC, SK Hynix, and Intel actively expand their NAND Flash bit output in 1Q21, the oversupply situation in the industry will become more severe, with a forecasted 6% QoQ increase in NAND Flash bit output and a 10-15% QoQ decline in NAND Flash ASP in 1Q21.

Susquehanna is a global trading firm which has various interests in silicon manufacturing - and part of that interest is naturally materialized in Intel. In a recent group call from the firm, some details on Intel's manufacturing and product design woes came to light, which point towards even more execution slips than we've already seen. During the call, a number of points were broached, including dismal yields for Intel's 10 nm manufacturing process as of its introduction in late 2018 (which is why it never saw mainstream adoption from the company). News that Intel is looking for a new CEO also don't instill confidence on current CEO Bob Swan's capacity to steer the Intel behemoth.

Improved yields on 10 nm are being reported due to deployment of Intel's SuperFin technology, which improved yields to upwards of 50%, but still keeps them under the ones achieved in Intel's 14 nm process; an eye-opening tidbit in that Cannon Lake on 10 nm originally saw yields of only 25% usable chips per wafer; and that backporting Rocket Lake meant Intel had to deal with unfathomably large chips and high power consumption characteristics. And to add insult to injury, there is still not a definite timetable for 7 nm deployment, with delays being expected to be worse than the previously reported 6-12 months. This all paints a somewhat grim picture for Intel's capacity to compete with TSMC-powered AMD in many of its most important markets; the blue giant won't topple, of course, but it's expected that five years from now, we'll be looking at a very different outlook in the market between AMD and Intel. You can check the talked-about points in the call via the transcript after the break. You should still take the transcript with a grain of salt.

Intel needs a platform refresh to battle the competition, mainly speaking to battle AMD and its Ryzen 5000 series processors. That is why the company is developing 500 series of chipsets covering the low-end (H510), mid-range (B560), and high-end markets (Z590) that pair with the upcoming Rocket Lake-S processor generation. Dubbed 11th generation of Core processors, the 11th generation of Intel Core CPUs are going to be built on Intel's refined 14 nm process. The CPUs are supposed to feature a Cypress Cove core, which is a backport of Golden Cove found in Ice Lake. The 500 series motherboards are the last in the DDR4 generation, launching in the timeframe when DDR5 is supposed to take over in the coming years.

Today, thanks to Weixin, a Chinese media outlet that posted a short story on the WeChat platform, we have information about the launch date of these new chipsets. According to the source, we are allegedly going to see these new chipsets on January 11th, the day that Intel CES 2021 event is supposed to happen. The platform will include a range of motherboards from Intel's partners and is supposed to bring support for the much-needed PCIe 4.0 protocol. The launch date should be taken with a grain of salt, of course, before taking it as a fact.

Today, at Intel's Memory and Storage 2020 event, the company highlighted six new memory and storage products to help customers meet the challenges of digital transformation. Key to advancing innovation across memory and storage, Intel announced two new additions to its Intel Optane Solid State Drive (SSD) Series: the Intel Optane SSD P5800X, the world's fastest data center SSD, and the Intel Optane Memory H20 for client, which features performance and mainstream productivity for gaming and content creation. Optane helps meet the needs of modern computing by bringing the memory closer to the CPU. The company also revealed its intent to deliver its 3rd generation of Intel Optane persistent memory (code-named "Crow Pass") for cloud and enterprise customers.

"Today is a key moment for our memory and storage journey. With the release of these new Optane products, we continue our innovation, strengthen our memory and storage portfolio, and enable our customers to better navigate the complexity of digital transformation. Optane products and technologies are becoming a mainstream element of business compute. And as a part of Intel, these leadership products are advancing our long-term growth priorities, including AI, 5G networking and the intelligent, autonomous edge." -Alper Ilkbahar, Intel vice president in the Data Platforms Group and general manager of the Intel Optane Group.

AMD is preparing to launch its next-generation of EPYC processors codenamed Millan. Being based on the company's latest Zen 3 cores, the new EPYC generation is going to deliver a massive IPC boost, spread across many cores. Models are supposed to range anywhere from 16 to 64 cores, to satisfy all of the demanding server workloads. Today, thanks to the leak from ExecutableFix on Twitter, we have the first benchmark of a system containing two of the 64 core, 128 thread Zen 3 based EPYC Milan processors. Running in the 2P configuration the processors achieved a maximum boost clock of 3.7 GHz, which is very high for a server CPU with that many cores.

The system was able to produce a Cinebench R23 score of insane 87878 points. With that many cores, it is no wonder how it is done, however, we need to look at how does it fare against the competition. For comparison, the Intel Xeon Platinum 8280L processor with its 28 cores and 56 threads that boost to 4.0 GHz can score up to 49,876 points. Of course, the scaling to that many cores may not work very well in this example application, so we have to wait and see how it performs in other workloads before jumping to any conclusions. The launch date is unknown for these processors, so we have to wait and report as more information appears.

According to the International Data Corporation (IDC) Worldwide Quarterly Server Tracker, vendor revenue in the worldwide server market grew 2.2% year over year to $22.6 billion during the third quarter of 2020 (3Q20). Worldwide server shipments declined 0.2% year over year to nearly 3.1 million units in 3Q20. Volume server revenue was up 5.8% to $19.0 billion, while midrange server revenue declined 13.9% to $2.6 billion, and high-end servers declined by 12.6% to $937 million.

"Global demand for enterprise servers was a bit muted during the third quarter of 2020 although we did see areas of strong demand," said Paul Maguranis, senior research analyst, Infrastructure Platforms and Technologies at IDC. "From a regional perspective, server revenue within China grew 14.2% year over year. And worldwide revenues for servers running AMD CPUs were up 112.4% year over year while ARM-based servers grew revenues 430.5% year over year, albeit on a very small base of revenue."

Amazon Web Services (AWS), the world's largest cloud services provider, has launched its Graviton series of custom processors some time ago. With Graviton, AWS had a plan to bring down the costs of offering some cloud services both for the customer and for the company. By doing that, the company planned to attract new customers offering greater value, and that plan seems to be working out well. When AWS launched its first-generation Graviton processor, the company took everyone by surprise and showed that it is capable of designing and operating its custom processors. The Graviton series of processors is based on the Arm Instruction Set Architecture (ISA) and the latest Graviton 2 series uses Arm Neoverse N1 cores as the base.

Today, thanks to the data from Liftr Insights, we get to see just how many total AWS instances are Graviton based. The data is showing some rather impressive numbers for the period from June 2019, to August 2020. In that timeframe, Intel with its Xeon offerings has seen its presence decrease from 88% to 70%, while AMD has grown from 11% to 20% presence. And perhaps the greatest silent winner here is the Graviton processor, which had massive growth. In the same period, AWS increased Graviton instance number from making up only 1% of all instances, to make up 10% of all instances available. This is a 10-fold increase which is not a small feat, given that data center providers are very difficult when it comes to changing platforms.

Today, thanks to the ServeTheHome forum member "111alan", we have the first pictures of the alleged Intel Sapphire Rapids Xeon processor. Pictured is what appears to be a dual-die design similar to Cascade Lake-SP design with 56 cores and 112 threads that uses two dies. The Sapphire Rapids is a 10 nm SuperFin design that allegedly comes even in the dual-die configuration. To host this processor, the motherboard needs an LGA4677 socket with 4677 pins present. The new LGA socket, along with the new 10 nm Sapphire Rapids Xeon processors are set for delivery in 2021 when Intel is expected to launch its new processors and their respective platforms.

The processor pictured is clearly a dual-die design, meaning that Intel used some of its Multi-Chip Package (MCM) technology that uses EMIB to interconnect the silicon using an active interposer. As a reminder, the new 10 nm Sapphire Rapids platform is supposed to bring many new features like a DDR5 memory controller paired with Intel's Data Streaming Accelerator (DSA); a brand new PCIe 5.0 standard protocol with a 32 GT/s data transfer rate, and a CXL 1.1 support for next-generation accelerators. The exact configuration of this processor is unknown, however, it is an engineering sample with a clock frequency of a modest 2.0 GHz.

Raja Koduri senior vice president, chief architect, and general manager of Architecture, Graphics, and Software at Intel Corporation has recently teased Intel's upcoming Xe-HP accelerator alongside its in production HC3 XG310 server card. The HC3 solution was Intel's first Xe-based product utilizing the Xe-LP architecture. The Intel Xe-LP products are Intel's lowest power efficiency optimized Xe processors while the Xe-HP products should offer improved performance and scaling. The upcoming Xe-HP accelerator appears to be a single-slot passively card with a single 8-pin power connector. Raja Koduri expects developers will begin receiving Xe-HP, Xe-HPG, and Xe-HPC products in 2021. He also declared that we are in the GPU golden age with new launches from NVIDIA, AMD, Intel, and Apple.

ID Cooling has announced their latest Mega Coolers, a new moniker and lineup that announces the coolers' compatibility with the latest, greatest, highest performance, and most power-hungry CPUs. The coolers are designed to accompany CPUs with TDPs up to 280 W. ID Cooling said these new mega coolers in the SE-207-series, which are basically tower coolers in design, will be hitting market just in time to address AMD's new-generation enthusiast grade Ryzen 5000-series 'Vermeer' and Intel's upcoming 11th Generation Core 'Rocket Lake' processors.

There are two separate coolers in the family; the SE-207-XT Black is aimed at AMD's Ryzen processors as well as Intel CPUs in LGA1200/115x/20xx form-factors; if you need some more cooling oomph, however, the SE-207-TRX Black is aimed at AMD's Ryzen Threadripper processor in TR4 packaging. Both coolers offer the same 280 W TDP dissipation capabilities, and are built in a twin-tower design with twin fans (700 RPM ~ 1800 RPM, up to 76.16 CFM air pressure, up to 35.2 dBA). In the case of the Threadripper-loving SE-207-TRX, a copper baseplate with direct-touch heatpipe technology removes heat from your CPU and passes it through the large, black, dual fin-stack arrays which are then cooled by two 120 mm fans. The E-207-XT Black, on the other hand, makes do with a solid copper baseplate absent of that same direct touch technology.

Apple's M1 SoC is possibly the year's biggest semiconductor success story, as the chip has helped Apple begin its transition away from Intel's x86 machine architecture, and create its own silicon that's optimized for its software and devices; much like its A-series SoCs powering iOS devices. The company now plans to scale up this silicon with a new 32-core version designed for high-performance Mac devices, such as the fastest MacBook Pro models; and possibly even iMac Pros and Mac Pros. The new silicon could debut in a new-generation Mac Pro in 2022. Bloomberg reports that the new silicon will allow this workstation to be half the size of the current-gen Mac Pro workstation in form, while letting Apple keep its generational performance growth trajectory.

In addition, Apple is reportedly developing a 16-core "big" + 4 "small" core version of the M1, which could power more middle-of-the-market Macs, such as the iMac desktop, and the bulk of the MacBook Pro lineup. The 16B+4s core chip could debut as early as Spring 2021. Elsewhere, the company is reportedly stepping up efforts to develop its own high-end professional-visualization GPU that it can use in its iMac Pro and Mac Pro workstations, replacing the AMD Radeon Pro solutions found in the current generation. This graphics architecture will be built from the ground-up for the Metal 3D graphics API, as well as a parallel compute accelerator. Perhaps the 2022 debut of the Arm-powered Mac Pro could feature this GPU.

Today, in the latest GeekBench 5 submission by ASUS, we have discovered something rather interesting. Intel's Ice Lake-SP processors were rumored to arrive with up to 28 cores and 56 threads at maximum, on a single chip. That was due to the 10 nm process used to make these chips, with suspicions that the yield of the node was not good enough to make any higher core count parts. Thanks to the GB5 listing, discovered by Leakbench on Twitter, the Intel Ice Lake-SP CPU engineering sample appeared with an amazing 36 cores with 72 threads. This is supposedly Intel's efforts to try and match the 64 cores and 128 threads of AMD's EPYC "Rome" CPUs, which are winning many server applications due to their performance.

The 36C/72T chip was paired with another similar chip in a 2P dual-socket configuration, which made the total core count rise to 72 cores and 144 threads, running inside of Asustek's Y4R-A1-ASUS-G1 server. The system was reporting a clock frequency of 3.6 GHz base speed, which means that the possible boost clocks could be higher. The CPU features a 1.25 MB level two (L2) cache per core (45 MB in total) and 54 MB of unified level three (L3) cache. That makes this CPU core quite an improvement over the past Cooper Lake generation. We are waiting for more information about these CPUs, and we are going to report on it in the coming time.

In a bid to maximize utilization of its own semiconductor foundry for manufacturing larger, more profitable processors, Intel could be look at contracting TSMC to manufacture certain processors based on its low-power CPU microarchitectures, according to a new Intel job posting discovered by Komachi Ensaka. The job description for a position in Intel's Bengaluru facility, speaks of a "QAT Design Integration Engineer" who would play a role in the "development and integration of CPM into Atom and Xeon-based SoC on Intel and TSMC process."

QAT is a hardware feature that accelerates cryptography and data-compression workloads. Since the Xeon part in this sentence is referenced next to SoC, Intel could be referring to Xeon processors based on low-power cores, such as "Snow Ridge," which uses "Tremont" CPU cores. The decision to go with TSMC could also be driven by the 5G infrastructure hardware gold rush awaiting the likes of Intel across dozens of new markets, particularly those averse to buying hardware from Huawei.

During CES 2020, way back in January of this year, Razer had shown off a quite interesting concept. Called a modular gaming desktop, the concept has a goal to allow users to just swap-out parts on the fly and have no trouble doing so. Today, the company has officially decided to launch the Tomahawk gaming desktop. Designed for small-form-factor computing, the case of the Tomahawk PC is coming in at just 10L volume, with measurements of 210 mm x 365 mm x 150 mm. The case is an all-black aluminium silhouette with the signature Razer logo and Chroma lighting around the base. That gives it a simple look that can blend in with any environment.

When it comes to the insides, the PC features a power supply of 750 Watts that powers one of Intel's NUC Element boards that is a house for a 45 W Core i9-9980HK Coffee Lake processor with eight cores and 16 threads. When it comes to memory, it has 16 GB of RAM and 512 GB of PCIe M.2 NVMe SSD storage, paired with a 2 TB hard drive. Razer offers users to upgrade memory and storage, while the CPU is soldered to the board. You can pre-order the Razer Tomahawk PC at a price starting at $2,399.99, while if you want to equip it with something like NVIDIA GeForce RTX 3080 GPU, you will be paying $3,199.99. If you already have a GPU to install, then you should just order the base.

At an Intel Labs virtual event today, Intel unveiled Horse Ridge II, its second-generation cryogenic control chip, marking another milestone in the company's progress toward overcoming scalability, one of quantum computing's biggest hurdles. Building on innovations in the first-generation Horse Ridge controller introduced in 2019, Horse Ridge II supports enhanced capabilities and higher levels of integration for elegant control of the quantum system. New features include the ability to manipulate and read qubit states and control the potential of several gates required to entangle multiple qubits.

"With Horse Ridge II, Intel continues to lead innovation in the field of quantum cryogenic controls, drawing from our deep interdisciplinary expertise bench across the Integrated Circuit design, Labs and Technology Development teams. We believe that increasing the number of qubits without addressing the resulting wiring complexities is akin to owning a sports car, but constantly being stuck in traffic. Horse Ridge II further streamlines quantum circuit controls, and we expect this progress to deliver increased fidelity and decreased power output, bringing us one step closer toward the development of a 'traffic-free' integrated quantum circuit."-Jim Clarke, Intel director of Quantum Hardware, Components Research Group, Intel.

Today, at Intel Labs Day, Intel highlighted industry-leading technological advances toward the realization of the company's long-standing vision of integrating photonics with low-cost, high-volume silicon. The advancements represent critical progress in the field of optical interconnects, which address growing challenges around the performance scaling of electrical input/output (I/O) as compute-hungry data workloads increasingly overwhelm network traffic in data centers. Intel demonstrated advances in key technology building blocks, including miniaturization, paving the way for tighter integration of optical and silicon technologies.

"We are approaching an I/O power wall and an I/O bandwidth gap that will dramatically hinder performance scaling. The rapid progress Intel is making in integrated photonics will enable the industry to fully re-imagine data center networks and architectures that are connected by light. We have now demonstrated all of the critical optical technology building blocks on one silicon platform, tightly integrated with CMOS silicon. Our research on tightly integrating photonics with CMOS silicon can systematically eliminate barriers across cost, power and size constraints to bring the transformative power of optical interconnects to server packages." -James Jaussi, senior principal engineer and director of PHY Lab, Intel Labs.

An alleged Ashes of the Singularity (AotS) benchmark results page for the top 11th Gen Core "Rocket Lake" processor leaked to the web courtesy TUM_APISAK. It's official now that Intel will keep its lengthy processor model number schemes, with the top part being the Core i9-11900K, a successor to the i9-10900K. It also confirms that the "Rocket Lake" silicon caps out at 8-core/16-thread, with performance on virtue of the IPC gains from the new "Cypress Cove" CPU cores."Cypress Cove" is believed to be a back-port of "Willow Cove" to the 14 nm silicon fabrication process that "Rocket Lake-S" is built on.

The screenshot also confirms the nominal clocks (base frequency) of the i9-11900K to be 3.50 GHz, as Intel tends to put base frequency in the name-string of its processors. Paired with a GeForce RTX 3080 and 32 GB of RAM, the i9-11900K-powered machine yielded 62.7 FPS CPU frame-rate at 1440p resolution, and 64.7 FPS CPU frame-rate at 1080p (a mere 3.18% drop in frame-rates from the increase in resolution). These numbers put the i9-11900K in the same league as the Ryzen 7 5800X in CPU frame-rates tested under similar conditions.

Today at AWS re:Invent 2020, AWS CEO Andy Jassy announced EC2 instances that will leverage up to eight Habana Gaudi accelerators and deliver up to 40% better price performance than current graphics processing unit-based EC2 instances for machine learning workloads. Gaudi accelerators are specifically designed for training deep learning models for workloads that include natural language processing, object detection and machine learning training, classification, recommendation and personalization.

"We are proud that AWS has chosen Habana Gaudi processors for its forthcoming EC2 training instances. The Habana team looks forward to our continued collaboration with AWS to deliver on a roadmap that will provide customers with continuity and advances over time." -David Dahan, chief executive officer at Habana Labs, an Intel Company.

When AMD introduced its Smart Access Memory technology, everyone was wondering will other GPU and CPU providers, namely Intel and NVIDIA, develop a similar solution to complement their offerings. The SAM technology is just AMD's way of naming PCIe resizable Base Address Register (BAR) technology, which has been present in PCI specifications for years as an optional feature. Why it's emerging now you might wonder. Well, the currently used PCIe revision has reached enough bandwidth on the bus to complement the complex data movement that GPU requires and now supports the use of the wider VRAM frame buffer.

It appears that not only AMD has this technology in its portfolio. ASUS has updated its BIOS firmware for its ROG Maximus XII Apex motherboard based on the Intel Z490 chipset, with some pretty interesting features. According to Tom's Hardware, we have information that the next release of BIOS firmware update 1003 for the ROG Maximus XII Apex motherboard will bring support for resizable BAR, making it a first on an Intel platform. For now, the beta 1002 BIOS supports it, however, a stable version will roll out in BIOS 1003. With the motherboard using PCIe 3.0 standard, a lower-bandwidth revision compared to AMD's platform, it will be interesting to see how resizable BAR is performing once the first tests come.

Update 09:45 am UTC: Chris Wefers, ASUS PR Germany, has announced that resizable BAR will be coming to all ASUS motherboards with Intel Z490/H470/B460 chipset, with alleged 13.37% performance increase in Forza Horizon 4, per ASUSes testing. You can see the test configuration in the image below.

As a world leader in computer memory, TEAMGROUP understands the importance of getting ahead in the next generation of DDR technology, hence it will be releasing ELITE series DDR5 memory in 2021. With over 20 years of experience developing DDR3 and DDR4 products, the company has dazzled the world with its advanced R&D capabilities and excellent product quality. After the JEDEC announced the DDR5 memory standard, TEAMGROUP has been actively designing and working together with our IC manufacturing partners to pioneer and prepare for this new generation.

TEAMGROUP is leading the way with its first DDR5 memory under its global top-selling ELITE memory product line. It plans to release a 16 GB 4800 MHz module operating at 1.1 V, down from the 1.2 V of the previous generation. The data transfer rate is increased to 4,800-5,200 Mbps, an increase of up to 1.6 times while reducing power consumption by 10%. Today's DDR4 memory with error correction code (ECC) requires an additional chip installed on the PCB, whereas DDR5 supports on-die ECC, a feature that self-corrects single-bit errors, greatly improving system stability. Anticipation is high for the efficiency improvements brought by the new generation, which can be utilized for big data and AI computing and other related applications.

Total NAND Flash revenue reached US$14.5 billion in 3Q20, a 0.3% increase QoQ, while total NAND Flash bit shipment rose by 9% QoQ, but the ASP fell by 9% QoQ, according to TrendForce's latest investigations. The market situation in 3Q20 can be attributed to the rising demand from the consumer electronics end as well as the recovering smartphone demand before the year-end peak sales season. Notably, in the PC market, the rise of distance education contributed to the growing number and scale of Chromebook tenders, but the increase in the demand for Chromebook devices has not led to a significant increase in NAND Flash consumption because storage capacity is rather limited for this kind of notebook computer. Moreover, clients in the server and data center segments had aggressively stocked up on components and server barebones during 2Q20 due to worries about the impact of the pandemic on the supply chain. Hence, their inventories reached a fairly high level by 3Q20. Clients are now under pressure to control and reduce their inventories during this second half of the year. With them scaling back procurement, the overall NAND Flash demand has also weakened, leading to a downward turn in the contract prices of most NAND Flash products.

AMD's main competitor to Intel's 11th Gen Core "Tiger Lake" processor in the mobile space, the Ryzen 5800U, will introduce the same kind of generational IPC improvements over the Ryzen 4800U "Renoir" as the Ryzen 5000 desktop processors introduced over their Ryzen 3000 predecessors. Based on the 7 nm "Cezanne" silicon, the new Ryzen 7 5800U processor was put through Geekbench 5.1.1, where it yielded performance numbers of 1491 points single-threaded, and 6450 points multi-threaded. HotHardware comments that these numbers reflect a major IPC increase.

With the Ryzen 5000U series, AMD is taking a very confusing approach to the processor model stack, with half the parts based on the older "Zen 2" microarchitecture and "Lucienne" silicon, and the other half "Zen 3." The model number scheme goes as 5x00U, where if "x" is an odd number, the chip is "Zen 2" based, and if it's an even number, it is "Zen 3" based. For example, the 5800U is based on "Zen 3," whereas the 5700U is based on "Zen 2." Find the 5800U Geekbench 5 validation here. The Geekbench database listing also confirms that much like with the 8-core "Zen 3" chiplets on the Ryzen 5000 "Vermeer" desktop processors, "Cezanne" features an 8-core "Zen 3" CPU that does away with the 4-core CCX arrangement, and features a single 8-core CCX with a monolithic 16 MB L3 cache—a doubling in overall L3 cache amount compared to "Renoir," and a quadrupling in addressable L3 cache by each core.

ASRock Rack, a branch of ASRock focused on making server products, has today launched a new motherboard that can accommodate up to 64 core AMD EPYC CPU. Built on the new, proprietary form factor called "Deep Mini-ITX", the ROMED4ID-2T motherboard is just a bit bigger than the standard ITX board. The standard ITX boards are 170 x 170 mm, while this Deep Mini-ITX form extends the board by a bit. It measures 170 x 208.28 mm, or 6.7" x 8.2" for all of the American readers. ASRock specifies that the board supports AMD's second-generation EPYC "Rome" 7002 series processors. Of course, the socket for these CPUs is socket SP3 (LGA4094) with 4094 pins.

The motherboard comes with 4 DDR4 DIMM slots, of any type. Supported DIMM types are R-DIMM, LR-DIMM, and NV-DIMM. If you want the best capacity, LR-DIMM use enables you to use up to 256 GB of memory. When it comes to expansion, you can hook-up any PCIe 4.0 device to the PCIe 4.0 x16 slot. There is also an M.2 2280 key present, so you can fit in one of those high-speed PCIe 4.0 x4 M.2 SSDs. For connection to the outside world, the board uses an Intel X550-AT2 controller that controls two RJ45 10 GbE connectors. There are also two Slimline (PCIe 4.0 x8 or 8 SATA 6 Gb/s), and four Slimline (PCIe 4.0 x8) storage U.2 ports.