AMD Radeon RX 9070 XT and RX 9070 "MBA" models have turned up for sale through Chinese black market channels; with local PC hardware enthusiasts quickly snapping up these "reference design" curios—for "scientific" analysis, and in some cases: bragging rights. Officially, Team Red has relied on its board partners to produce an all-custom first wave of RDNA 4 gaming graphics cards. Recent discoveries of authentic-looking "Made by AMD" specimens suggested a very limited distribution of first-party units to trusted partners. According to a fresh ITHome news article, Team Red's Radeon RX 9070 XT reference PCB design was fully revealed via a comprehensive teardown. Kyogre shared heavily watermarked photos with the online publication; showcasing their disassembly of a "leaked" example.

Team Red's proprietary triple-fan cooling solution and backplate were removed; leading to the exposure of gory innards. ITHome provided a quick rundown visual clues—the analyzed unit features: "dual PCIe 8-Pin power supply interfaces, and a black shell with a tough and simple shape. It is about three slots thick and is equipped with a rear open and ventilated cooling backplane with a core back frame...Looking deeper into the internal structure of the graphics card, we can find that this graphics card is equipped with five heat pipes, the corresponding position of the GPU core uses a copper base, and the surrounding memory positions are also in contact with the metal frame through thermal pads. In addition, it is equipped with four video output interfaces."

G.SKILL International Enterprise Co., Ltd., the world's leading brand of performance overclock memory and PC components, is thrilled to announce the introduction of an enhanced revision of DDR5 R-DIMM memory to the market. Based on the newly released JEDEC revision standards for DDR5 R-DIMM, the new G.SKILL R-DIMM memory features a 16-layer PCB for improved signal integrity, as well as adding two transient voltage suppression (TVS) diodes and a fuse for voltage protection. Designed for enterprise servers and professional workstations, the new DDR5 R-DIMM memory brings cutting-edge advancements to performance and protection.

Improved Signal Integrity with 16-Layer PCB
These new DDR5 R-DIMM memory modules feature an advanced 16-layer PCB, a significant upgrade from previous 8- or 10-layer designs. The increase in the PCB layer count improves signal integrity for better reliability and stability in data transmission, even under high-performance or overclocked workloads - making it an ideal DDR5 R-DIMM memory choice for high-performance computing applications on servers or workstations.

German overclocking specialist Der8auer has showcased a custom water block for NVIDIA's flagship RTX 5090 Founders Edition, achieving a remarkable 30-degree Celsius reduction in core GPU temperatures. The prototype copper cooling solution demonstrated peak operating temperatures of just 43.8°C, compared to the stock cooler's 73.8°C under identical testing conditions. The design features an all-copper construction with a substantial 14 mm-thick cold plate, engineered specifically for optimal thermal transfer from the GB202 die. While maintaining the dual-slot form factor of the original Founders Edition cooler, the water block incorporates multiple cooling channels beneath a transparent top panel that reveals the coolant flow.

Memory temperatures also significantly improved, dropping by 38 degrees Celsius below stock measurements. Despite its impressive thermal performance, the prototype faces several engineering challenges. The team was constrained by the graphics card's display output ribbon cable, which prevented a more compact design closer to the actual PCB dimensions. Additionally, the Founders Edition card's unique split PCIe finger design presented structural concerns that require additional reinforcement before any potential commercial release. Der8auer's prototype prioritizes function over aesthetics, omitting common features like RGB lighting in favor of raw cooling performance. The water block's industrial design reflects its proof-of-concept nature, though the transparent top panel offers practical utility for monitoring coolant flow. While no retail release timeline has been announced, Der8auer indicated that addressing the PCIe slot structural support issue remains a critical milestone before any commercial version could be considered.

ASUS China has launched an extensive customer support program addressing potential graphics card damage linked to its PCIe Q-Release Slim mechanism. The initiative includes full motherboard replacements, compensation packages, and warranty coverage for affected customers. The compensation package encompasses a 200 RMB ($27) store credit and complete motherboard replacement for affected units across multiple product lines, including the Z790, B860, Z890, B850, and X870 series motherboards. Particularly, the affected motherboard lineup includes high-end models like the ROG Maximus Z890 Extreme and ROG Crosshair X870E Hero. The controversy centers on the mechanism's internal peg design, which reportedly can affect graphics card PCIe connectors during repeated installation and removal cycles.

ASUS China has confirmed the development of a revised Q-Release Slim design, though no specific timeline has been announced. Warranty coverage varies by graphics card manufacturer. ASUS and MSI have committed to priority repairs and free card replacements for both PCB wear and gold finger damage when repairs prove unsuccessful. However, Gigabyte and Colorful cards require evaluation at repair centers before determining coverage eligibility. This regional response contrasts sharply with ASUS's US position, which maintains that internal testing shows no correlation between the Q-Release Slim mechanism and graphics card damage. While ASUS continues to defend the mechanism's design integrity internationally, its comprehensive response in the Chinese market suggests a more cautious approach to potential hardware compatibility issues. We must wait and see if the worldwide ASUS position will change in the coming days and if the North American department issues any different response.

Yes, the title is correct. One of NVIDIA's GeForce RTX 5090D "China" edition GPUs, not the regular RTX 5090, managed to run at 3.4 GHz under liquid nitrogen. With a staggering 575 W default TDP, Tony Yu, ASUS China's general manager, has performed physical modifications that allow the card to run up to 1000 W TDP. The RTX 5090D is a China-exclusive variant with virtually no difference from the regular RTX 5090, just limited general AI capability due to US export regulations. ASUS China used its top-end Astral OC variant for this stunt, which, as we proved in our review of the regular ASUS RTX 5090 Astral OC, has some pretty good chip binning, allowing the card to reach the highest overclock. We pushed the regular RTX 5090 Astral OC GPU on air to 3086 MHz, a +277 MHz over the stock boost setting. However, the RTX 5090D equivalent under LN2 manages to reach 3,390 MHz at peak loads, which is a +581 MHz difference.

For memory, the overclock is equally impressive with 34 Gbps. Regarding performance, the LN2-overclocked RTX 5090D surpassed stock performance by approximately 16%. During benchmark tests, the GPU outperformed multiple previous-generation graphics cards, including a dual RTX 3090 Ti configuration in Port Royal and a quad GTX 1080 Ti setup in Fire Strike. Power consumption figures indicate that 1,760 W was used in total for a rig with ASUS ROG Astral RTX 5090D, which is paired with the Ryzen 7 9800X3D on the ASUS ROG X870E Hero motherboard. This roughly yields a 1,000 W power consumption by the card, which has seen its PCB get physical modifications to output such high power.

A fake AMD Ryzen 7 9800X3D processor was recently discovered through AMD's after-sales service inspection in China after a customer reported their CPU wouldn't boot. The counterfeit unit revealed several technical discrepancies that distinguish it from genuine processors, though these differences are difficult to spot without detailed examination. While authentic Ryzen 7 9800X3D processors use PCB number 33050, the counterfeit unit featured PCB number 32546—a designation typically associated with Ryzen 7000 series R7 or R9 processors. This suggests the counterfeiters may be repurposing older generation chips or PCBs to create fake 9000 series processors. The physical appearance of the counterfeit unit showed additional inconsistencies.

Genuine Ryzen 9000 series processors feature a distinctive blue PCB, while the fake unit displayed a dark green coloration characteristic of older models. Component patches adjacent to the PCB also failed to match the standard configuration of the Ryzen 9000 series. In a notable error, some counterfeit units were mislabeled as "Ryzen 9 9800X3D"—a model number that doesn't exist in AMD's product lineup. For comparison, legitimate Ryzen 7 9800X3D units can be identified by their blue-tinted PCB, while older models like the R9 7900X and R7 7700X feature dark green PCBs. The information was initially shared on Chiphell by sources within the PCB manufacturing industry, who provided side-by-side comparisons of authentic and counterfeit units. AMD's after-sales service has confirmed that processors purchased outside official distribution channels will not receive warranty support or service protection. Buyers should buy CPUs only thought official sale distributors to avoid such cases.

MSI has provided a detailed look into its motherboard production facilities through a new interactive webpage. The company has outlined the 15-step manufacturing process that transforms bare circuit boards into complete motherboards. The process begins with solder paste printing, where a precise layer of conductive paste is applied to the bare PCB. This is followed by automated inspection using imaging systems to verify proper paste placement. High-speed machines then position electronic components onto the board before they move through reflow ovens that permanently bond the components. Multiple inspection stages are integrated throughout the assembly line, including automated optical systems that check both sides of the board. Specialized machines handle routing, component insertion, wave soldering, and alloy welding.

Each board undergoes circuit testing and receives mechanical components through automated screw-locking systems before final functional testing and molding. The complexity of modern motherboard manufacturing requires rigorous quality checks and final power tests before the motherboard leaves manufacturing facilities. For interested enthusiasts, MSI's new landing page is here. The manufacturing reveal accompanies MSI's launch of two new motherboard series—Z790 and B860. To celebrate this launch, MSI has announced promotional events running from January 20 to March 31, 2025, including Steam game codes with select product purchases. A new reward program offers points for product reviews and referrals.

Details of ASUS' next generation BTF 2.0 (BTF stands for Back to the Future) platform have appeared online, courtesy of Andreas Schilling from HardwareLuxx. The editor posted pictures from an ASUS presentation over at Bluesky that details one major change to the power connector design, which will make the BTF 2.0 graphics cards compatible with non BTF systems. This suggests that ASUS might consider offering more graphics cards with BTF 2.0 support, since they can be used in any system with the proposed changes.

As you may or may not know, the first generation BTF graphics cards ended up with a power connector at the rear, bottom edge of the graphics cards, placed somewhat behind the PCIe gold fingers. The cards also lacked any other power connectors, which made them exclusive for BTF builds. To solve this conundrum, ASUS moved its GC-HPWR PCB power connector up on the PCB, making sure it clears any motherboard components. However, instead of increasing the height of the motherboard BTF 2.0 connector, the company provides a small adapter that plugs in between the motherboard and graphics card, to make its new BTF 2.0 cards compatible with any system.

YUNZII has introduced its latest mechanical keyboard, the AL65 QMK/VIA aluminium keyboard, at CES 2025. The AL65 features a CNC-anodized aluminium frame and a distinctive hollow side-cut design that enhances both its aesthetics and RGB lighting effects. The keyboard is equipped with south-facing LEDs capable of displaying 16.8 million colors, with transparent PC keycaps to amplify the lighting. It supports QMK/VIA, allowing users to remap keys, create macros, and adjust lighting on both Windows and macOS systems. The AL65 offers tri-mode connectivity with wired, Bluetooth, and 2.4 GHz wireless options, powered by an 8000mAh battery for extended use. Its gasket-mounted structure reduces typing noise and provides a softer feel while sound-dampening foam layers minimize vibrations. A hot-swappable PCB supports both 3-pin and 5-pin switches, enabling users to customize the keyboard with different mechanical switches easily. With its robust build, versatile connectivity, and customizable features, the AL65 is designed to cater to a wide range of users, from typists to gamers.

NVIDIA's GeForce RTX 5090 graphics card printed circuit board has allegedly been shown in the flesh, showing the memory layout and some interesting engineering choices. The custom PCB variant (non-Founders Edition) houses more than 40 capacitors, which is perhaps not standard on the FE reference board, and 16 GDDR7 memory modules. The leaked PCB, which extends beyond standard dimensions and traditional display connector configurations, is reportedly based on NVIDIA's PG145 reference design. The memory modules are distributed systematically: five on the left, two below, five on the right, and four above the GPU die. The interface is PCIe 5.0 x16.

As NVIDIA has reportedly designated 32 GB GDDR7 memory capacity for these cards, this roughly translates into 16 x 2 GB GDDR7 memory modules. At the heart of the card lies what sources claim to be the GB202 GPU, measuring 24×31 mm within a 63×56 mm package. Power delivery uses a 16-pin 12V-6x2 power connector, as expected. The entire PCB features only a single power connector, so the 16-pin 12V-2x6, but with an updated PCIe 6.0 CEM specification, is the logical choice.

Thermal Grizzly's Thermal Putty offers a premium alternative to traditional thermal pads. It is electrically non-conductive, easy to apply, and functions as a flexible gap filler that compensates for height differences. This makes it an ideal replacement for thermal pads in graphics cards. Graphics cards are typically equipped with thermal pads of varying heights from the factory. When replacing these pads or upgrading to a GPU water cooler, matching replacement pads are necessary.

TG Thermal Putty can compensate for height differences from 0.2 to 3.0 mm, making it a versatile solution. Thermal Putty can be applied in two ways. Firstly, it can be applied over large areas using the included spatulas. Alternatively, it can be applied manually (gloves are recommended). When applied by hand, small beads can be shaped to fit the specific contact surfaces (e.g., VRAM, SMD).

Both pictures and videos of a partial teardown of Apple's recently launched Mac mini with the M4 SoC have appeared online courtesy of various Chinese sources. There are at least two interesting parts to these partial teardowns and they're related to storage and WiFi. On the storage front, Apple has moved away from having soldered NAND chips straight on the main PCB of the Mac mini, to instead having them on a custom PCB which is similar to M.2, but a custom Apple design. The PCB pictured contained a pair of 128 GB NAND chips and with the source of the teardown being from China, there's also a video showing a repair shop desoldering the two chips and replacing them with two 1 TB chips, or in other words, the SSD was upgraded from 256 GB to 2 TB.

The upgrade brought with it some extra performance as well, even if the write speed remained at a comparatively slow 2900 MB/s, the read speed went up from 2000 MB/s to 3300 MB/s which is a significant gain in performance. This is obviously not a consumer friendly upgrade path, but we'd expect to see third party upgrade options at some point in the future, assuming there's no black listing of third party storage modules. The NAND controller is still likely to be integrated into Apple's SoC, but the PCB that the NAND flash chips are mounted onto appears to have some kind of SPI flash on it as well, which might make third party upgrades a lot harder.

Gigabyte, the world's leading computer brand, proudly announces that the AORUS Z890 series motherboards are now officially available for purchase. Designed to maximize the performance of the latest Intel Core Ultra processors, the groundbreaking D5 Bionic Corsa technology was introduced along with advanced thermal management and optimized power design on these boards. GIGABYTE's continuous partnership with HWiNFO further enhances the boards with real-time monitoring of CPU vCore power phase outputs and efficiency. With these powerful boards now on sale, users can enjoy unmatched performance and seamless customization, making them the ideal platform for enthusiasts and professionals.

D5 Bionic Corsa is the core technology for AORUS Z890 series motherboards, which leverages AI-enhanced innovations across software, hardware, and firmware to boost DDR5 memory speeds to an unprecedented 9500+ MT/s. The AI SNATCH Engine, powered by advanced AI overclocking models, optimizes configurations for DDR5 XMP memory and CPUs, enabling up to 20% faster speeds. With XMP AI BOOST and CPU AI BOOST, users can achieve world-class overclocking performance with just one click. The AI-driven PCB Design enhances signal integrity by reducing reflection by 28.2%, while HyperTune BIOS fine-tunes the Memory Reference Code (MRC) for peak performance. The VRM Thermal Balance mechanism ensures heat dissipation across the VRM with a heatpipe design, while optimized PWM firmware balances current output for superior stability.

Intel today announced that its lead products on Intel 18A, Panther Lake (AI PC client processor) and Clearwater Forest (server processor), are out of the fab and have powered-on and booted operating systems. These milestones were achieved less than two quarters after tape-out, with both products on track to start production in 2025. The company also announced that the first external customer is expected to tape out on Intel 18A in the first half of next year.

"We are pioneering multiple systems foundry technologies for the AI era and delivering a full stack of innovation that's essential to the next generation of products for Intel and our foundry customers. We are encouraged by our progress and are working closely with customers to bring Intel 18A to market in 2025." -Kevin O'Buckley, Intel senior vice president and general manager of Foundry Services

It feels that only a few days have passed since we announced the HatDrive! Nano, and there is much more in the pipeline that we're excited to share with you! Today, though, we have our Pineboards Ai Bundle (Hailo 8L) to whet your ever-hungry AI appetites, and we think you're going to love it.

Combining an M.2 2280 M-Key NVMe connection with an M.2 2230 A/E-Key connection pre-loaded with a Hailo-8L on a bottom-mounted Raspberry Pi 5 HAT enables you to get your AI fix whilst also being able to boot and make use of fast NVMe storage. This builds on the success of our ever-popular Raspberry Pi 5 AI HAT and Google Coral combinations, but massively bumps the processing power, enabling you to do so much more!

Today, we have the latest round of leaks that suggest that AMD's upcoming RDNA 4 graphics cards, codenamed the "RX 8000-series," might continue to rely on GDDR6 memory modules. According to Kepler on X, the next-generation GPUs from AMD are expected to feature 18 Gbps GDDR6 memory, marking the fourth consecutive RDNA architecture to employ this memory standard. While GDDR6 may not offer the same bandwidth capabilities as the newer GDDR7 standard, this decision does not necessarily imply that RDNA 4 GPUs will be slow performers. AMD's choice to stick with GDDR6 is likely driven by factors such as meeting specific memory bandwidth requirements and cost optimization for PCB designs. However, if the rumor of 18 Gbps GDDR6 memory proves accurate, it would represent a slight step back from the 18-20 Gbps GDDR6 memory used in AMD's current RDNA 3 offerings, such as the RX 7900 XT and RX 7900 XTX GPUs.

AMD's first generation RDNA used GDDR6 with 12-14 Gbps speeds, RDNA 2 came with GDDR6 at 14-18 Gbps, and the current RDNA 3 used 18-20 Gbps GDDR6. Without an increment in memory generation, speeds should stay the same at 18 Gbps. However, it is crucial to remember that leaks should be treated with skepticism, as AMD's final memory choices for RDNA 4 could change before the official launch. The decision to use GDDR6 versus GDDR7 could have significant implications in the upcoming battle between AMD, NVIDIA, and Intel's next-generation GPU architectures. If AMD indeed opts for GDDR6 while NVIDIA pivots to GDDR7 for its "Blackwell" GPUs, it could create a disparity in memory bandwidth performance between the competing products. All three major GPU manufacturers—AMD, NVIDIA, and Intel with its "Battlemage" architecture—are expected to unveil their next-generation offerings in the fall of this year. As we approach these highly anticipated releases, more concrete details on specifications and performance capabilities will emerge, providing a clearer picture of the competitive landscape.

SPARKLE unveiled its low-profile series around mid-January—this lineup included an Intel Arc A380 GENIE dual-fan/dual-slot model and an Arc A310 ECO (single-slot, single-fan config) card. Compact device expert/YouTuber, ETA Prime, has uploaded a fascinating video that covers a modification project that involved a hybridization of SPARKLE's latest low-profile graphics cards and a Minisforum MS-01 test system. SPARKLE has released various models based on Intel's "Alchemist" Arc A380 6 GB GPU, but their PCB design is shared across a range of cooling options. ETA Prime could not source an aftermarket lower-profile cooler for his SPARKLE A380 GENIE, so he resorted to cannibalizing the A310 ECO model for relevant parts.

The ECO's single-slot cooling solution was not well proportioned enough to make contact with the SPARKLE A380 GENIE's VRM, so ETA Prime had to "add an aftermarket heatsink." He sold the remaining unneeded pieces—A310 board and GENIE cooler—to a friend for $60. The resultant hybrid—the "world's first-ever single-slot Intel Arc A380"—was bunged into the SFF Minisforum MS-01 test system. Notable specs included the Intel Core i9-13900H CPU and 32 GB of DDR5-5200 RAM. ETA Prime utilized Acer's Predator BiFrost graphics card utility to "trick" in a stable 54 W power limit. 60-ish FPS performance results—with low-to-medium settings at 1080p across a selection of games—were promising, especially for a restrictive small form factor build. ETA Prime hopes that SPARKLE will launch a smaller A380 model in the future—alternatively a specialist firm could produce a nice aftermarket copper part.

EMTEK products rarely pop up on TPU's news section, but the GPU database contains a smattering of the South Korean manufacturer's Ampere-based GeForce RTX graphics card. VideoCardz has discovered an updated MIRACLE X3 White model—EMTEK's latest release is a GeForce RTX 4070 SUPER 12 GB card. The triple-fan model seems to stick with NVIDIA's reference specifications—VideoCardz also noticed a physical similarity: "under the cooler shroud, the card boasts a non-standard U-shaped PCB, reminiscent of Team Green's Founders Edition. However, it remains uncertain whether EMTEK utilizes the same PCB as NVIDIA." The asking price—of ₩919,990—converts to around $680, when factoring in regional taxes. EMTEK's MIRACLE X3 cooling solution seems to be fairly robust—featuring four 6 mm heat pipes—so an adherence to stock clocks is a slight surprise. The company's GAMING PRO line includes a couple of factory overclocked options.

Altair, a global leader in computational intelligence, announced the upcoming release of Altair SimSolid for electronics, bringing game-changing fast, easy, and precise multi-physics scenario exploration for electronics, from chips, PCBs, and ICs to full system design. "As the electronics industry pushes the boundaries of complexity and miniaturization, engineers have struggled with simulations that often compromise on detail for expediency. Altair SimSolid will empower engineers to capture the intricate complexities of PCBs and ICs without simplification," said James R. Scapa, founder and chief executive officer, Altair. "Traditional simulation methods often require approximations when analyzing PCB structures due to their complexity. Altair SimSolid eliminates these approximations to run more accurate simulations for complex problems with vast dimensional disparities."

Altair SimSolid has revolutionized conventional analysis in its ability to accurately predict complex structural problems with blazing-fast speed while eliminating the complexity of laborious hours of modeling. It eliminates geometry simplification and meshing, the two most time-consuming and expertise-intensive tasks done in traditional finite element analysis. As a result, it delivers results in seconds to minutes—up to 25x faster than traditional finite element solvers—and effortlessly handles complex assemblies. Having experienced fast adoption in the aerospace and automotive industries, two sectors that typically experience challenges associated with massive structures, Altair SimSolid is poised to play a significant role in the electronics market. The initial release, expected in Q2 2024, will support structural and thermal analysis for PCBs and ICs with full electromagnetics analysis coming in a future release.

PowerColor, a leader in innovative graphics card solutions, is proud to announce the launch of its latest marvel in the gaming world: the PowerColor RX 7900 XT Hellhound - Spectral White Edition. Following the overwhelming success of the 7800 XT Spectral White model, this new release brings the same unique design elements to the powerful Radeon 7900 XT series.

Spectral White Edition: A Symphony in White
The RX 7900 XT Hellhound Spectral White Edition is a testament to PowerColor's commitment to exceptional design and engineering. It features an all-white PCB, heatsink, and cooling solution, setting a new benchmark in graphics card aesthetics. This card is not just a performance powerhouse; it's a statement piece for any gaming setup, offering a pristine, unified look that white PC build enthusiasts will adore.

The PC master race has yielded many interesting activities for enthusiasts alike, with perhaps the pinnacle of activities being overclocking. Usually, subjects for overclocking include CPUs, GPUs, and RAM, with other components not actually being capable of overclocking. However, the enthusiast force never seems to settle, and today, we have proof of overclocking an off-the-shelf 2.5-inch SATA III NAND Flash SSD thanks to Gabriel Ferraz, a Computer Engineering graduate, and TechPowerUp's SSD database maintainer. He uses the RZX Pro 256 GB SSD in the video, a generic NAND Flash drive. The RZX Pro uses the Silicon Motion SM2259XT2 single-core, 32-bit ARC CPU running up to 550 MHz. It has two channels at 400 MHz, each with eight chip enable interconnects, allowing up to 16 NAND Flash dies to operate. The SSD doesn't feature a DRAM cache or support a host memory buffer. It has only one NAND Flash memory chip from Kioxia, uses BiCS FLASH 4 architecture, has 96 layers, and has 256 GB capacity.

While this NAND Flash die is rated for up to 400 MHz or 800 MT/s, it only ran at less than half the speed at 193.75 MHz or 387.5 MT/s at default settings. Gabriel acquired a SATA III to USB 3.0 adapter with a JMS578 bridge chip to perform the overclock. This adapter allows hot swapping of SSDs without the need to turn off the PC. He shorted two terminals in the drive's PCB to get the SSD to operate without its default safe mode. Mass Production Tools (MPTools), which OEMs use to flash SSDs, were used to change the firmware settings. Each NAND Flash architecture has its own special version of MPTools. The software directly shows control of the Flash clock, CPU clock, and output driving. However, additional tweaks like Flash IO driving with subdivisions need modifications. Control and Flash On-Die Termination (ODT) and Schmitt window trigger (referring to the Schmitt trigger comparator circuit) also needed a few modifications to make it work.

Yeston, Chinese GPU AIB, has posted a preview of its upcoming Sakura RTX 40 SUPER cards, featuring a white PCB, a first for SUPER cards. The current RTX 40 Sakura models are set to be updated to these SUPER variants, with no reported visual changes. The RTX 4070 (Ti) SUPER will adopt the older design first introduced with the RTX 30 series, featuring a color scheme of white, purple, blue, and adorned with sakura-inspired graphic designs. In addition to the RTX 4070 (Ti) SUPER, Yeston will also release the RTX 4080 SUPER, which is expected to be the highest-end Sakura model from the company at the time of its release. The only RTX 4090 in Yeston's lineup is a Deluxe edition. These new releases offer gamers an exceptional GPU design, particularly those who appreciate the unique aesthetic of the Sakura series. The white PCB is a notable feature that sets these cards apart in the market. More details about the release date, clock speeds, and pricing are expected to be announced soon.

GIGABYTE TECHNOLOGY Co. Ltd, a leading manufacturer of premium gaming hardware, today launched GeForce RTX 40 SUPER series graphics cards powered by NVIDIA ADA Lovelace architecture. Featuring AORUS, AERO, GAMING, EAGLE and WINDFORCE series, the lineup completely serves the needs for every customer, from gamers, creators, to AI developers.

The new GeForce RTX SUPER GPUs are the ultimate way to experience AI on PCs. Specialized AI Tensor Cores deliver up to 836 AI TOPS to deliver transformative capabilities for AI in gaming, creating and everyday productivity. PC gamers demand the very best in visual quality, and AI-powered NVIDIA Deep Learning Super Sampling (DLSS) Super Resolution, Frame Generation and Ray Reconstruction combine with ray tracing to offer stunning worlds. With DLSS, seven out of eight pixels can be AI-generated, accelerating full ray tracing by up to 4x with better image quality.

The recent U.S. government restrictions on AI hardware exports to China have significantly impacted several key semiconductor players, including NVIDIA, AMD, and Intel, restricting them from selling high-performance AI chips to Chinese land. This ban has notably affected NVIDIA's GeForce RTX 4090 gaming GPUs, pushing them out of mainland China due to their high computational capabilities. In anticipation of these restrictions, NVIDIA reportedly moved a substantial inventory of its AD102 GPUs and GeForce RTX 4090 graphics cards to China, which we reported earlier. This could have contributed to the global RTX 4090 shortage, driving the prices of these cards up to 2000 USD. In an interesting turn of events, insiders from the Chinese Baidu forums have disclosed that specialized factories across China are repurposing these GPUs, which arrived before the ban, into AI solutions.

This transformation involves disassembling the gaming GPUs, removing the cooling systems and extracting the AD102 GPU and GDDR6X memory from the main PCBs. These components are then re-soldered onto a domestically manufactured "reference" PCB, better suited for AI applications, and equipped with dual-slot blower-style coolers designed for server environments. The third-party coolers that these GPUs come with are 3-4 slots in size, whereas the blower-style cooler is only two slots wide, and many of them can be placed in parallel in an AI server. After rigorous testing, these reconfigured RTX 4090 AI solutions are supplied to Chinese companies running AI workloads. This adaptation process has resulted in an influx of RTX 4090 coolers and bare PCBs into the Chinese reseller market at markedly low prices, given that the primary GPU and memory components have been removed.Below, you can see the dismantling of AIB GPUs before getting turned into blower-style AI server-friendly graphics cards.

Just over a month ago, Gigabyte launched a pair of white motherboards for Intel CPUs and now the company has quietly released a pair of white models for AMD CPUs. The two models are the B650 AORUS Elite AX Ice and the B650M AORUS Elite AX Ice, the main difference being the form factor, as the latter is a microATX board. Besides the white PCB, all of the heatsinks are silver/white in colour, although all the connectors on the boards are black, just as with the Intel boards. The rear I/O shield has also been given the white treatment to better fit into a white case.

However, the Ice boards are not identical in terms of specs to their older siblings, as Gigabyte has done a few minor changes that aren't noticeable at first look. Gigabyte has shifted away from using MTK/AMD and Intel WiFi cards and is now using the Realtek RTL8852CE on these new boards. We should mention that Gigabyte are using the RTL8852CE on later revisions of the black versions as well. It's currently unknown when these new boards will be available and how much they'll retail for.