BIOSTAR, a leading manufacturer of motherboards, graphics cards, and storage devices today, unveils the brand new TZ590-BTC motherboard. Designed based on the Intel Z590 chipset that runs Intel 10th and 11th gen processors, BIOSTAR's new TZ590-BTC motherboard targets the mid-range audience of casual gamers, content creators, and users looking to build an HTPC for casual-content consumption. The TZ590-BTC also specializes in crypto mining with native inbuilt functionality to support this claim.

Available in a full-ATX form factor, the new TZ590-BTC motherboard provides a strong feature-rich base for users, with functionality such as PCIe 4.0, PCIe M.2 4.0. An 11 Phase power design controlled by BIOSTAR's signature DIGITAL PWM technology keeps everything running smooth and safe. 4x DDR4 RAM slots that support up to 3600 (OC) with a maximum capacity of 128 GB come in handy for all purposes. The TZ590-BTC has 8x PCIe 3.0x1 Slots, ideal for BITCOIN and ETHEREUM mining, as well as a single PCIe 4.0 x16 Slot targeting exceptional performance.

TerraMaster, a professional brand that specializes in providing innovative storage products for home, businesses and enterprises, introduces the all-new T9-423 9-bay high-performance NAS designed for small- and medium-sized businesses. The TerraMaster T9-423 is ideal for enterprise applications such as file storage servers, mail servers, web servers, FTP servers, virtual machine hosts, and other enterprise-level applications.

The TerraMaster T9-423 is powered by a quad-core 2.00 GHz (2.90 GHz boost) Intel Core processor and comes with two 2.5 GbE ports with support for Link Aggregation, delivering speeds of up to 283 MB/s. It supports multi-user and high-concurrent file access. Its compact chassis can be mounted vertically and horizontally to provide builders flexibility of use.

Cincoze's latest DI-1100 series of compact, high-performance, and power-saving rugged embedded industrial computers with rich interface options is receiving a warm welcome in the smart factory industry. The DI-1100 series is ideal for autonomous mobile robots (AMRs), ensuring on-time delivery of components to the production line or goods to warehouse racks, and suitable for 24/7 operation in outdoor remote monitoring systems. These diverse applications are crucial to all smart factories looking to reduce personnel, control costs, increase shipments, and implement outdoor monitoring.

Structurally, there are three major parts to an AMR: the hardware, the software, and the user interface. The robot's hardware consists of the mechanical design, sensors, motion control, and the computer. The computer is the most critical part for integration because it must serve as the robot's center and interface with various devices and software interfaces. For this application, it is essential to choose a high-performance, power-saving, and compact rugged industrial embedded computer with diverse I/O. Cincoze's latest DI-1100 series uses a high-performance 8th Gen quad-core Intel Core i7/i5/i3 (Whiskey Lake-U) CPU and supports one DDR4 2400 MHz SO-DIMM up to 64 GB to meet the high-efficiency computing requirements for real-time dynamic detection, obstacle detection, and multitasking.

Adobe's Premiere Pro, one of the most common video editing tools in the industry, has received a February update today with version 22.2. The new version brings a wide array of features like Adobe Remix, an advanced audio retiming tool. Alongside that, the latest update accelerates offline text-to-speech capabilities by as much as three times. However, this is not the most significant feature, as we are about to see. Adobe also enabled 10-bit 420 HDR HEVC H/W encoding on Window with Intel and NVIDIA graphics. This feature allows the software to use advanced hardware built-in the NVIDIA Quadro RTX and Intel Iris Xe graphics cards.

The company managed to run some preliminary tests, and you can see the charts below. They significantly improve export times with the latest 22.2 software version that enables HEVC 10-Bit hardware encoding. For Intel GPUs, no special drivers need to be installed. However, for NVIDIA GPUs, Adobe is advising official Studio drivers in combination with Quadro RTX GPUs.

Despite its technical advantages over NAND flash, Intel's Optane memory technology never really took off in the consumer space, largely due to its much higher costs. Based on details from Intel's SEC filings, it would appear that its corporate customers haven't been overly impressed either, as Intel filed a US$576 million loss in 2020 with regards to its Optane business.

As to if this trend continued in 2021, we're going to have to wait and see, but it looks like the losses are set to continue based on its revenue figures, according to Blocks and Files. With this in mind, it seems like Micron's exit from 3D XPoint memory isn't so hard to understand, as the technology is clearly too costly to make up for the benefits on offer, for most of its potential customer base. Transitions to higher-density 3D XPoint memory has also been slow, which might be yet another reason customers have shied away from using it, even as a caching solution in storage servers. It wouldn't surprise us if Intel moves on from Optane/3D XPoint memory in the near future, as the company seems to have shifted its focus firmly away from storage solutions.

Ansys today announced its inaugural partnership to IFS Accelerator - EDA Alliance to provide best-in-class EDA tools and simulation solutions that will support customer innovation, including bespoke silicon for customizable three-dimensional integrated circuit (3D-IC) designs. By leveraging Ansys' market-leading multiphysics solutions, IFS Accelerator will make silicon technology available to customers to design uniquely innovative chips. Ansys' cutting-edge EDA and simulation tools will enable mutual customers to reduce design barriers, minimize design risk and cost, and accelerate time-to-market.

The IFS Accelerator will foster collaborative innovation with world-leading EDA, design services and IP partners to provide a comprehensive design ecosystem with premium process technologies, advanced packaging technologies, and manufacturing capabilities. "We are excited to announce the IFS Accelerator - EDA Alliance as a major step forward for Intel's foundry ambitions," said Rahul Goyal, VP and GM of Intel Product & Design Ecosystem Enablement. "Together with Ansys and other partners, this alliance will create advanced flows and methodologies, and accelerate productivity by combining our knowledge, resources, and shared passion to drive electronic design."

The world's leading computer brand, GIGABYTE's new generation AERO creator series laptops featuring Intel's latest 12 generation processors and RTX 30 series graphics cards have hit shelves. The dual chip approach improved processing performance by 28% vastly improving the time efficiency for creators. The newly released AERO 16 utilizes a 16:10 golden ratio 4K OLED screen and a 4 side super narrow bezel design revolutionizing the field of vision for creators. In addition to boasting a film industry standard of 100% DCI-P3 wide color gamut, each laptop has undergone correction and certification by color authorities, X-Rite and Pantone, before leaving the factory ensuring the most accurate color display.

The AERO 16, designed specifically for creators, also captures the essence of creators with an aesthetic, chassis carefully integrated with CNC aluminium alloy, every detail skillfully crafted by GIGABYTE designers. Coming along with the AERO 16 and AERO 17, the AERO HUB is a must-have mobile workstation for content creators.

Intel put out a clear, close-up picture of its Arc "Alchemist" gaming graphics card engineering sample. This matches with a picture of the PCB rear shot that surfaced in a report by "Moore's Law is Dead." The picture reveals a PCB that's about 3/4th the length of the cooling solution, with the remainder of the cooler's length being used to directly vent airflow from the back.

The PCB reveals a rectangular pad for the GPU, which corresponds with that of the larger "Alchemist" GPU. This is surrounded by what look like eight GDDR6 memory pads for a 256-bit wide memory interface; at least 10 VRM phases of an unknown configuration; and a power input configuration that's made up of one each of 8-pin and 6-pin PCIe power connectors (capable of delivering 300 W including slot power). The PCB shows traces that connect the GPU to all 16 PCI-Express lanes of the PCIe finger. Display outputs include three full-size DisplayPorts and an HDMI. This particular variant of "Alchemist" is rumored to feature 512 execution units (4,096 unified shaders), and at least in SiSoft SANDRA, it allegedly outperforms the GeForce RTX 3070 Ti "Ampere."

The European High Performance Computing Joint Undertaking (EuroHPC JU) has launched 3 new research and innovation projects. The projects aim to bring the EU and its partners in the EuroHPC JU closer to developing independent microprocessor and HPC technology and advance a sovereign European HPC ecosystem. The European Processor Initiative (EPI SGA2), The European PILOT and the European Pilot for Exascale (EUPEX) are interlinked projects and an important milestone towards a more autonomous European supply chain for digital technologies and specifically HPC.

With joint investments of €140 million from the European Union (EU) and the EuroHPC JU Participating States, the three projects will carry out research and innovation activities to contribute to the overarching goal of securing European autonomy and sovereignty in HPC components and technologies, especially in anticipation of the European exascale supercomputers.

Mesh shader is a relatively new concept of a programmable geometric shading pipeline, which promises to simplify the whole graphics rendering pipeline organization. NVIDIA introduced this concept with Turing back in 2018, and AMD joined with RDNA2. Today, thanks to the finds of Phoronix, we have gathered information that Intel's DG2 GPU will carry support for mesh shaders and bring it under Vulkan API. For starters, the difference between mesh/task and traditional graphics rendering pipeline is that the mesh edition is much simpler and offers higher scalability, bandwidth reduction, and greater flexibility in the design of mesh topology and graphics work. In Vulkan, the current mesh shader state is NVIDIA's contribution called the VK_NV_mesh_shader extension. The below docs explain it in greater detail:

Vulkan API documentationThis extension provides a new mechanism allowing applications to generate collections of geometric primitives via programmable mesh shading. It is an alternative to the existing programmable primitive shading pipeline, which relied on generating input primitives by a fixed function assembler as well as fixed function vertex fetch.

There are new programmable shader types—the task and mesh shader—to generate these collections to be processed by fixed-function primitive assembly and rasterization logic. When task and mesh shaders are dispatched, they replace the core pre-rasterization stages, including vertex array attribute fetching, vertex shader processing, tessellation, and geometry shader processing.

Intel is expanding its Bug Bounty program with Project Circuit Breaker, bringing together a community of elite hackers to hunt bugs in firmware, hypervisors, GPUs, chipsets and more. Project Circuit Breaker broadens and deepens Intel's existing open Bug Bounty program by hosting targeted time-boxed events on specific new platforms and technologies, providing training and creating opportunities for more hands-on collaboration with Intel engineers. Project Circuit Breaker's first event, Camping with Tigers, is already underway with a group of 20 researchers who received systems with Intel Core i7 processors (formerly "Tiger Lake").

Project Circuit Breaker is possible thanks to our cutting-edge research community. This program is part of our effort to meet security researchers where they are and create more meaningful engagement. We invest in and host bug bounty programs because they attract new perspectives on how to challenge emerging security threats - and Project Circuit Breaker is the next step in collaborating with researchers to strengthen the industry's security assurance practices, especially when it comes to hardware. We look forward to seeing how the program will evolve and to introducing new voices to the meaningful work that we do."
-Katie Noble, director, Intel Product Security Incident Response Team (PSIRT) and Bug Bounty

Intel's Alder Lake processors have two types of cores present, with two distinct sets of features and capabilities enabled. For example, smaller E-cores don't support the execution of AVX-512 instructions, while the bigger P-cores have support for AVX-512 instructions. So Intel has decided to remove support for it altogether not to create software errors and run into issues with executing AVX-512 code on Alder Lake processors. This happened just months before the launch of Alder Lake, making us see some initial motherboard BIOSes come with AVX-512 enabled from the box. Later on, all motherboard makers pulled the plug on it, and it is a rare sight to see support for it.

However, it seems like MSI is unhappy with the lack of AVX-512, and the company is reenabling partial support for it. According to Xaver Amberger, editor at Igor's Lab, MSI reintroduces selecting microcode version with its MEG Z690 Unify-X motherboard. There is an option for AVX-512 enablement in the menu, and it is indeed a functional one. With BIOS A22, MSI enabled AVX-512 instruction execution, and there are benchmarks to prove it works. This shows an advantage of 512-bit wide execution units of AVX-512 over something like AVX2, which offers only 256-bit wide execution units. In applications such as Y-Cruncher, AVX-512 enabled the CPU to reach higher performance targets while consuming less power.

The world's leading computer brand, GIGABYTE, launched its new generation of AORUS professional gaming laptops featuring Intel's 12th generation processors and Nvidia's RTX 30Ti series graphics cards. The upgraded dual-chip approach has resulted in a great leap in performance and smoother gaming experience, with a 11% increase in gaming performance. In addition to an upgraded performance, AORUS has pushed the limits of size dimensions of notebooks in order to improve visual enjoyment for gamers.

The AORUS 17, GIGABYTE's flagship model laptop debuted at CES 2022, changed the rules of the game by introducing a 17-inch screen panel inside a 15-inch tall laptop chassis. By adopting an extremely narrow four-sided frame design, AORUS increased the screen-to-body ratio to 90%, so you see more while carrying less. Without compromising any performance, the screen still boasts a 360 Hz refresh rate, allowing players to see more frames and win more games.

Global semiconductor sales revenue grew by a more than comfortable 25% in 2021, hitting a new record above $500 Billion for the first time. The data, part of Gartner's preliminary report on the state of the industry in 2021, bookends yet another year plagued with shortages and too little supply for the thirst of the consumer and business sectors. Naturally, supply constraints have led to higher ASP (Average Selling Prices) for the hardware that does get manufactured and distributed, as we've seen all too well in the graphics card market.

"As the global economy bounced back in 2021, shortages appeared throughout the semiconductor supply chain, particularly in the automotive industry," said Andrew Norwood, research vice president at Gartner. "The resulting combination of strong demand as well as logistics and raw material price increases drove semiconductors' average selling price higher (ASP), contributing to overall revenue growth in 2021.

3DMark Fire Strike Hall of Fame is where overclockers submit their best hardware benchmark trials and try to beat the very best. For years, one record managed to hold, and today it just got defeated. According to an extreme overclocker called "biso biso" from South Korea and a part of the EVGA OC team, the top spot now belongs to the AMD Radeon RX 6900 XT graphics card. The previous 3DMark Fire Strike world record was set on April 22nd in 2020, when Vince Lucido, also known as K|NGP|N, set a record with four-way SLI of NVIDIA GeForce GTX 1080 Ti GPUs. However, that record is old news since January 27th, when biso biso set the history with AMD Radeon RX 6900 XT GPU.

The overclocker scored 62389 points, just 1,183 more from the previous record. He pushed the Navi 21 XTX silicon that powers the Radeon RX 6900 XT card to an impressive 3,147 MHz. Paired with a GPU memory clock of 2370 MHz, the GPU was probably LN2 cooled to achieve these results. The overclocker used EVGA's Z690 DARK KINGPIN motherboard with Intel Core i9-12900K processor as a platform of choice to achieve this record. You can check it out on the 3DMark Fire Strike Hall of Fame website to see yourself.

Raja Koduri, Intel's Senior Vice President and General Manager for the Accelerated Computing Systems and Graphics Group, has responded via Twitter to an open letter from PC Gamer that addressed the nightmare-like state for GPU pricing and availability, classifying it as "a huge issue for PC gamers and the industry at large". Intel says they are looking to put millions of Arc GPUs into PC Gamer's hands as the company fast approaches the end of the reaffirmed Q1 launch window for its new high-performance discrete graphics products. The reading on this is that Intel plans to add to the available mass of GPUs that consumers can buy, thus alleviating the strain from overwhelming consumer demand, and bringing about a much healthier market - with real and acceptable pricing for graphics products.

That Intel plans to ship millions of Arc GPUs to consumers is no surprise; the company definitely wants to recoup its investment in developing consumer-oriented, high-performance graphics architectures. However, any claims or expectations of improved GPU supply in the market should be taken with a grain of salt, as the bottleneck for graphics products stands not at the GPU design level, but at the semiconductor manufacturing one: namely, there are only so many GPU wafers that graphics chips designers can secure from foundry company TSMC, which also serves customers like Apple, Qualcomm, and other technology industry giants.

MAINGEAR—an award-winning PC system integrator of custom gaming desktops, notebooks, and workstations—today announced that new NVIDIA GeForce RTX 3050 graphics cards are now available to configure within MAINGEAR's product line of award-winning custom gaming desktop PCs and workstations. Featuring support for real-time ray tracing effects and AI technologies, MAINGEAR PCs equipped with the NVIDIA GeForce RTX 3050 offer gamers next-generation ray-traced graphics and performance comparable to the latest consoles.

Powered by Ampere, the NVIDIA GeForce RTX 3050 features NVIDIA's 2nd gen Ray Tracing Cores and 3rd generation Tensor Cores. Combined with new streaming multiprocessors and high-speed G6 memory, the NVIDIA GeForce RTX 3050 can power the latest and greatest games. NVIDIA RTX on 30 Series GPUs deliver real-time ray tracing effects—including shadows, reflections, and Ambient Occlusion (AO). The groundbreaking NVIDIA DLSS (Deep Learning Super Sampling) 2.0 AI technology utilizes Tensor Core AI processors to boost frame rates while producing sharp, uncompromised visual fidelity comparable to high native resolutions.

Intel Corporation today reported fourth-quarter and full-year 2021 financial results. The company also announced that its board of directors approved a cash dividend increase of five percent to $1.46 per share on an annual basis. The board declared a quarterly dividend of $0.365 per share on the company's common stock, which will be payable on March 1 to shareholders of record as of February 7.

"Q4 represented a great finish to a great year. We exceeded top-line quarterly guidance by over $1 billion and delivered the best quarterly and full-year revenue in the company's history," said Pat Gelsinger, Intel CEO. "Our disciplined focus on execution across technology development, manufacturing, and our traditional and emerging businesses is reflected in our results. We remain committed to driving long-term, sustainable growth as we relentlessly execute our IDM 2.0 strategy."

Remember that €1.06 billion antitrust lawsuit that Intel was slapped with by the European Commission back in 2009? It's ok if you don't, but it involved Intel being accused of "market malpractice, by influencing computer hardware manufacturers to postpone and/or cancel launches of their products that use CPUs made by its rival AMD" based on our own reporting from 2009. As these thing goes, Intel appealed to a higher court and that higher court handed back the case to the lower court who has now withdrawn the fine and the judges went as far as to say "The (European) Commission's analysis is incomplete and does not make it possible to establish to the requisite legal standard that the rebates at issue were capable of having, or likely to have, anticompetitive effects,".

The European Commission has said it will study the new judgement and will consider its next steps, but it seems unlikely that they'll be able to bring this case to court again, without some additional proof of wrongdoing. The end result of this is also likely to make it tougher to bring cases like to court in the future for the European Commission, as they will have to provide more detailed cases where they prove that things like MDF and rebates to their customers cause real, anti-competitive damages to other companies in the same line of business, in this case AMD. The case can still be appealed to the CJEU, so this 13 year saga might still not be over.

G.SKILL International Enterprise Co., Ltd., the world's leading manufacturer of extreme performance memory and gaming peripherals, is thrilled to announce the achievement of a new overclocking world record for fastest memory frequency at DDR5-8888 CL88-88-88-88, in cooperation with ASUS. This amazing frequency speed was achieved by the extreme overclocker "lupin_no_musume" with G.SKILL Trident Z5 DDR5 memory, ASUS ROG Maximus Z690 APEX motherboard, and Intel Core i9-12900K processor. To see the moment this amazing overclocking world record was set, please click the following video link: https://youtu.be/OgQFbUOs6i8

DDR5-8888 CL88-88-88-88 - Pushing the Speed to the Limit
At the dawn of the DDR5 era, G.SKILL and ASUS have been constantly exploring the memory speed limitations of the latest Intel Z690 platform. Surpassing the previous DDR5-8704 world record in November 2021, a new memory frequency world record is achieved at DDR5-8888 under liquid nitrogen extreme cooling. The memory speed has been validated by CPU-Z. Please refer to the screenshot and validation link below: https://valid.x86.fr/qgvylc

You may, or may not have noticed that in certain parts of the interweb, groups of people that are generally referred to as "Overclockers" have managed to get their cheap Celeron G6900's and Core i3-12100's to run at much higher clock speeds than Intel intended and now the company is unhappy about it, as they're anticipating that they're going to lose sales of more expensive CPUs. As such, Intel has issued a warning via Tom's Hardware
"Intel's 12th Gen non-K processors were not designed for overclocking. Intel does not warranty the operation of processors beyond their specifications. Altering clock frequency or voltage may damage or reduce the useful life of the processor and other system components, and may reduce system stability and performance."

Jokes aside, the lower end SKU's of Intel's 12th gen Alder Lake CPUs seem to be phenomenal overclockers, if you have the right motherboards. If the motherboard doesn't have an external clock gen, plus support for adjusting the BCLK on non-K CPUs, then you're not going to have much luck. This means, at least at the moment, that you're looking at fairly pricey Z690 motherboard, although there are rumors that we can expect the odd B660 motherboard that will get an external clock gen, with at least three models already reported to have BCLK adjustment support via beta UEFI updates. Pro Overclockers have already managed to hit speeds in excess of 5.3 GHz with the Celeron G6900 and that is only by adjusting the BCLK and the Voltage, which is no mean feat, as the CPU has fixed clock speed of 3.4 GHz, which makes this a 57 percent boost in clock speed. Intel is said to be looking into this unintended ability to overclock these CPU SKUs and is apparently looking at locking down this ability with a new microcode update in a future UEFI release.

Update: Added a screenshot from TPU's upcoming Core i3-12100F review, showing 5.2 GHz at 130 MHz BCLK.

Intel's upcoming Arc Alchemist lineup of discrete graphics cards generates a lot of attention from consumers. Leaks of these cards' performance and detailed specifications appear more and more as we enter the countdown to the launch day, which is sometime in Q1 of this year. Today, we managed to see a slide from @9950pro on Twitter that shows the laptop memory configuration of Intel's DG2 GPU. As the picture suggests, we can see that the top-end SKU1 with 512 EUs supports a 16 GB capacity of GDDR6 memory that runs at 16 Gbps speeds. The memory runs on a 256-bit bus and generates 512 GB/s bandwidth while having eight VRAM modules present.

When it comes to SKU2, which is a variant with 384 EUs, this configuration supports six VRAM modules on a 192-bit bus, running at 16 Gbps speeds. They generate a total capacity of 12 GBs and a bandwidth of 384 GB/s. We have SKU3 DG2 GPU going down the stack, featuring 256 EUs, four VRAM modules on a 128-bit bus, 8 GB capacity, and a 256 GB/s bandwidth. And last but not least, the smallest DG2 variants come in the form of SKU4 and SKU5, feating 128 EUs and 96 EUs, respectively. Intel envisions these lower-end SKUs with two VRAM modules on a 64-bit bus, and this time slower GDDR6 memory running at 14 Gbps. They are paired with 4 GB of total capacity, and the total bandwidth comes down to 112 GB/s.

Not all products are launched with an accompanying press release or even at a trade show and ASUS' new PL63 mini PC is one of those products. It might not be the most unique mini PC on the market, but as far as we're aware, it's the first with Thunderbolt 4 support, that is as long as you get the right SKU, as most seem to support USB 3.2 instead. The PL63 is built around Intel's 11th gen mobile processors and the three CPU options available are the Core i3-1115G4, Core i5-1135G7 and the Core i7-1165G7, all with Iris Xe graphics..

Other features include support for two SO-DIMMs, one PCIe 3.0 NVMe drive, 2.5 Gbps Ethernet, an HDMI 2.0 port, DP 1.4 as well as DP output through both the USB-C ports, regardless of if you get the USB 3.2 or Thunderbolt 4 version. As per the pictures, WiFi is also supported, with various options depending on the SKU. An external 19 V power brick rated at 90 W is included to power the 166.2 x 119.7 x 33.9 mm (WxDxH) system. Pricing for a barebone model seems to be US$508 for the Core i5-1135G7 based model, although it's unclear if this is with or without Thunderbolt 4 support. Bump that up to US$650 and you get 4 GB of RAM, a 128 GB SSD and Windows 10 Pro included.

Rather unusually, Intel announced its latest chip fab plans not via a press release, but via an exclusive article in TIME magazine. It seems like an unusual strategy, as these kinds of things are normally not announced in this kind of fashion, especially as TIME doesn't exactly have close ties with Intel, nor the semiconductor industry as a whole, but maybe this is Intel's new way of trying to change the image of the company. Either which way, Intel is apparently planning on building no less than two fabs on the 1,000 acre (~4 square kilometer) site in New Albany, Ohio, which should be the workplace of some 3,000 people once it stands ready in 2025.

The article quotes Pat Gelsinger saying "Our expectation is that this becomes the largest silicon manufacturing location on the planet," as Intel has the option to double the land for its new fab site and apparently has plans for as many as eight fabs at the location. Additional fabs obviously depends on demand and crucially if Intel manages to full off its contract foundry business, since without it, it seems unlikely that Intel is going to need the additional six fabs in the foreseeable future, especially as Intel is in the final stages of finishing its new fab in Ireland, while also planning to announce a location for yet another fab somewhere in Europe and let's not forget Arizona. The TIME article goes into a lot more details as to what the new fabs mean for Ohio, but doesn't go into much more detail about Intel's plans for its future fabs there.

Update: Official press release below.

Intel's Arc Alchemist discrete lineup of graphics cards is scheduled for launch this quarter. We are getting some performance benchmarks of the DG2-512EU silicon, representing the top-end Xe-HPG configuration. Thanks to a discovery of a famous hardware leaker TUM_APISAK, we have a measurement performed in the SiSoftware database that shows Intel's Arc Alchemist GPU with 4096 cores and, according to the report from the benchmark, just 12.8 GB of GDDR6 VRAM. This is just an error on the report, as this GPU SKU should be coupled with 16 GB of GDDR6 VRAM. The card was reportedly running at 2.1 GHz frequency. However, we don't know if this represents base or boost speeds.

When it comes to actual performance, the DG2-512EU GPU managed to score 9017.52 Mpix/s, while something like NVIDIA GeForce RTX 3070 Ti managed to get 8369.51 Mpix/s in the same test group. Comparing these two cards in floating-point operations, Intel has an advantage in half-float, double-float, and quad-float tests, while NVIDIA manages to hold the single-float crown. This represents a 7% advantage for Intel's GPU, meaning that Arc Alchemist has the potential for standing up against NVIDIA's offerings.