Maxsun has shared full specifications for the SK-PTLNAS and SK-WCLAIO, two NAS-focused MoDT (Mobile on Desktop) motherboards built around Intel's Panther Lake and Wildcat Lake chips. Both were first shown at Computex 2026 and are expected to launch this month, though pricing is still unknown. The SK-PTLNAS now tops out at the Core Ultra 9 386H, a 16-core chip with four Xe3 GPU cores, a 25 W base power and up to 80 W of turbo power. The Computex listing mentioned options up to the Core Ultra X9 388H, which no longer seems to be on the table. It supports up to 128 GB of DDR5-7200 memory across two SO-DIMM slots, with five M.2 slots and three Mini SAS ports for storage. The board also gets a PCIe 5.0 x8 slot, 10GbE, 2.5GbE and two 1GbE ports, plus two Thunderbolt 4 ports. Maxsun lists Windows 11 and Ubuntu support.

The SK-WCLAIO measures just 90 x 165 mm, a big change from the mATX design shown in June. It runs the six-core Core 7 350 (2P + 4LPE) with two Xe3 cores, a single DDR5-6400 SO-DIMM slot for up to 48 GB, 1 TB of onboard UFS 3.1 storage and four M.2 slots. Wildcat Lake has very few PCIe lanes, so those drives probably won't each get full x4 bandwidth. Maxsun rates the system at 45 W when active and 16 W with drives on standby, which should suit an always-on home NAS. The power input is a bit unclear for now, since the listing mentions both a 12 V DC jack and USB-C PD.

NVIDIA has officially implemented changes across its "Vera Rubin" product line, reducing target specifications for memory and interconnect fabric. According to the updated product specification sheet, the first change involves HBM4 memory. Previously, NVIDIA used HBM4 with a speed of 11.7 Gbps per pin, which has now been reduced to 9.6 Gbps per pin. In a "Vera Rubin" NVL72 rack-scale solution, NVIDIA now offers a total memory bandwidth of 1,400 TB/s, down from the previously advertised 1,580 TB/s. While this issue does not stem directly from NVIDIA's suppliers, it indicates that the company's implementation of memory bus width and overall electrical signaling is insufficient to support higher speeds.

However, the changes are not limited to memory. NVIDIA has also reduced the previously advertised NVLink bandwidth per GPU, decreasing it from 1.8 TB/s to 1.5 TB/s. This results in a reduction from 3.6 TB/s to 3 TB/s in bidirectional NVLink GPU bandwidth. The underlying issue is related to signal integrity, as NVIDIA reportedly lowered the NVLink Nyquist frequency target from 56 GHz to 47 GHz. Lowering the Nyquist frequency reduces the electrical demands on the SerDes, cables, and materials, which can improve manufacturing yields and overall feasibility, even though it requires accepting a small reduction in peak interconnect bandwidth. Thus, NVIDIA chose to sacrifice some bandwidth to enhance system stability. Below is the new system specification in light mode, and old system specification in dark mode.

Microsoft might be working on a cheaper 12-inch Surface, and Apple's MacBook Neo looks like the reason. DigiTimes reports, citing supply chain sources, that Microsoft is reviewing the specifications of its entry-level 12-inch Surface to bring costs down, and is also rethinking which processor goes inside. The report's headline, spotted by harukaze5719 on X, goes as far as saying Apple is forcing Microsoft to downgrade the device, and hints at a comeback for AMD with an Arm chip. Reportedly, the processor, memory, camera, battery, and chassis are all being looked at, though nothing seems to be finalized yet.

The pricing gap between the two laptops is quite large. Microsoft only just refreshed the 12-inch Surface Pro with Snapdragon X2 Plus, and it now starts at $1,149, up from $799 last year, partly because the 8 GB base configuration was dropped. Apple launched the MacBook Neo at $599, and even after a $100 increase in June that Apple blamed on memory costs, it still sits at $699. As for the chip, AMD's "Sound Wave" Arm APU was previously linked to Microsoft's 2026 Surface lineup and has even shown up in shipping manifests, so it looks like a candidate. Leaks put it at 2 performance and 4 efficiency Arm cores, RDNA 3.5 graphics with 4 CUs, and a 5 W to 10 W power target. Then again, Microsoft could just as well stick with Qualcomm, and neither Microsoft nor AMD has confirmed anything.

Montage Technology announced the mass-production of its DDR5 Clock Driver Plus (CK01P), designed to deliver stable and reliable clocking for next-generation client memory modules.

Compliant with the JEDEC DDR5CKD01 clock driver specification, the CK01P primarily targets next-generation client memory modules, including DDR5 CUDIMM, CSODIMM, CQDIMM, and CAMM and supports data rates of up to 9200 MT/s. The CK01P supports three clock operating modes—single PLL, dual PLL, and PLL bypass—to accommodate a wide range of system requirements. Featuring a zero-phase-delay clock architecture, the device supports programmable per-channel delay and allows output characteristics to be configured through control registers, enabling flexible adaptation to different DIMM network topologies. In addition, unused clock outputs can be disabled to further reduce power consumption.

Certified Ultra96 HDMI cables, required to unlock the full HDMI 2.2 specification, are now available for retail purchase. VideoCardz spotted Silkland taking orders for its new H96 series, which appears to be among the first certified Ultra96 cables available to consumers, offering up to 96 Gbps of bandwidth. ELECOM also revealed its own Ultra96 cables on September 8, advertised as the first certified products of their type in Japan, though those aren't expected to ship until mid-November. Silkland's cables, by comparison, opened for orders on September 14. Currently, there's nothing on the market that can actually use that bandwidth, so these cables are targeted towards future products.

HDMI 2.2 was finalized back in June 2025, and it enables some genuinely big jumps on paper, like uncompressed 8K at 60 Hz or 4K at 240 Hz with full 10-bit and 12-bit color. It also edges out DisplayPort 2.1's 80 Gbps ceiling. Actual HDMI 2.2 hardware, meaning GPUs, TVs, and monitors with the right silicon, isn't expected until 2027 at the earliest. Even then, AMD's upcoming RDNA 5 GPUs reportedly won't hit the full 96 Gbps tier, apparently topping out closer to 64 or 80 Gbps instead.

Micron is reportedly exploring the idea of developing high-endurance NAND Flash modules that would be positioned closer to the GPU, rather than being located far away in the storage pool behind multiple protocols. The company is said to be investigating ways to bring NAND Flash closer to handle workloads that don't require the latency and bandwidth of traditional HBM or regular DRAM modules. This architecture, currently referred to as "near-GPU NAND," would serve as a middle ground between density and durability. This concept would allow GPUs to access a pool of NAND Flash that has a lower density than conventional TLC or QLC NAND, but instead focuses on improving I/O speeds, bandwidth, and read times.

For instance, this would enable a GPU to access a potential storage pool with hundreds of gigabytes of storage directly on the GPU package or PCB, similar to what DRAM like HBM or GDDR7/LPDDR6 already does. This tier would sit between GPU memory and regular system storage, acting as a very fast buffer, which would be ideal for the inference of massive LLMs. At that point, LLMs wouldn't be memory-bound and could run on systems with fewer GPUs, but with more storage and memory tiers to compensate for the required inference size. Additionally, this NAND would be significantly cheaper than HBM or regular DRAM and would be sized according to system specifications. Images below are HBF slides from SK hynix and Sandisk.

Samsung Electronics today announced a new 14-inch Galaxy Book6, expanding the Galaxy Book6 lineup with a more accessible model designed for everyday computing. Starting at $799, Galaxy Book6 combines up to 25 hours of battery life, Intel Core 3 and Core 5 performance, Galaxy AI-enabled features, and the design language of the Galaxy Book6 family.

"With the new Galaxy Book6, we wanted to make the Galaxy Book experience available for more people," said Mincheol Lee, EVP, Head of Galaxy Ecosystems Business Team at Samsung Electronics. "At a more accessible price, it brings together the everyday essentials users value, including dependable performance, long-lasting battery life, connected Galaxy experiences, and Galaxy Book6's design DNA, giving users confidence that lasts beyond the purchase."

Chinese memory giant CXMT has reportedly begun risk production of HBM3E memory, according to The Information. This indicates that Chinese memory manufacturers are currently two generations behind South Korean companies, which are gradually transitioning to HBM4E memory production. However, this is a significant milestone for CXMT, which is on track to achieve a manufacturing capacity of around 350,000 DRAM wafers per month by the end of the year, with plans to increase that capacity in the coming years. It is reported that HBM3E manufacturing has started in very low quantities, signifying the commencement of risk production.

While we don't know the specifications of CXMT's HBM3E memory, we know that the standard JEDEC specification states that this must be a 1,024-bit width, and speeds of anywhere from 9.2 Gbps per pin to 12.4 Gbps, with most of the standard configurations targeting 9.6 Gbps per pin. The total bandwidth is around 1.2 TB/s, and capacities come in anywhere from 24 GB using 8-hi stacks or 36 GB using 12-high stacks of 24 Gb layers. However, given that CXMT progresses rapidly from risk production to mass production, we could see high volume manufacturing take place in just a few weeks.

CD Projekt Red announced The Witcher 3: Wild Hunt Remastered at Gamescom 2026, revealing a visual, performance, and gameplay upgrades coming as a free update to The Witcher 3 for existing owners. Owners of The Witcher 3: Wild Hunt Complete Edition will see their game update to the Remaster on September 29, but that upgrade will come with significantly higher demands of their gaming rigs, with the minimum PC system requirements now exceeding the recommended PC system requirements in the CPU department and more or less matching them on the GPU side of things.

According to the Remaster's Battle.Net store page, where the updated minimum specs were revealed first, The Witcher 3: Wild Hunt Remastered will call for an Intel Core i5-8400 or an AMD Ryzen 5 2600, paired with 12 GB of RAM and 70 GB of SSD storage. Previously, The Witcher 3: Wild Hunt only called for an Intel Core i5-2500K or an AMD A10-5800K.

G.SKILL has begun distributing payments between $20 and $25 to various customers after settling a $2.4 million class-action lawsuit. The lawsuit alleged that G.SKILL misled buyers of DDR4 and DDR5 memory kits about the actual speed of their DRAM kits. Filed in 2022, the lawsuit claimed that G.SKILL did not properly label its advertising materials and misled users into believing that the maximum speeds advertised for their memory modules were achievable only through XMP or EXPO memory overclocking profiles. Additionally, the lawsuit asserted that there were no warnings for consumers about the potential risks of memory overclocking, such as system instability or memory damage.

The DDR4 and DDR5 memory kits in the Trident Z and Ripjaws series were the focus of the class-action lawsuit. G.SKILL initially argued that it had properly labeled these products but eventually reached a settlement earlier this year, agreeing to pay $2.4 million in damages to the buyers of the original kits. Anyone who purchased G.SKILL DDR4 and DDR5 memory kits with speeds rated over 2,133 MT/s or 4,800 MT/s, respectively, from January 31, 2018, to January 7, 2026, is eligible to receive a settlement.

Noctua produces a range of cooling solutions, from low-profile air coolers and massive dual-tower air coolers to AIO liquid coolers. When fitting a large tower air cooler inside a case, PC case manufacturers list the maximum cooler height their chassis can accommodate. Recent testing by Noctua has revealed that more than half of the tested cases deviate from their original specifications. The room for error varies from being 3.5 mm less than the specified CPU cooler height to providing up to an extra 10 mm of cooler height. After testing over 100 cases, Noctua found that around 56 cases deviate from the manufacturer-provided specifications, mostly offering more space than advertised.

This is good news for case buyers, as there can be up to 10 mm more room for a CPU cooler. However, some cases are advertised with more space than they actually provide. For example, the ASUS Prime AP201 case claims to have 170 mm of CPU cooler headroom, but real-world measurements show it to be 168.3 mm. This means that when using the Noctua NH-D15 G2 dual-tower cooler, there's no way to increase RAM clearance by moving the front fan, necessitating the use of RAM kits that are only 32 mm tall. Anything taller will interfere with the CPU cooler. Another example is the Lian Li B4-mATX, which measured at 164.5 mm instead of the advertised 168 mm, making the NH-D15 G2 incompatible with this case.

Huawei has just unveiled its latest MateBook Pro laptop, featuring a homegrown processor, operating system, and a fully integrated experience. Priced at 9,999 RMB for the base model with 24 GB of RAM and a 512 GB SSD, this equates to about $1,480. The top-end configuration offers 32 GB of RAM and a 1 TB SSD for 13,999 RMB, approximately $2,072 at the current exchange rate.

Inside, the previous Kirin X90 has been replaced by the new Kirin X90 Plus, which is kept cool by a foam-copper vapor chamber, allowing the chip to maintain a 40 W performance envelope. Huawei seems to have retained the same core counts and GPU specs with a 10-core, 20-thread design, but the improved thermal management suggests they are extracting more performance from this generation. No official specifications have been released yet, so this remains speculative.

NVM Express, Inc. today announced the release of all 11 of the NVM Express (NVMe) specifications, incorporating new features and engineering change notices (ECNs). These updates represent the next step in the evolution of NVMe technology, with security, management and sustainability extensions to address the growing demands of data-intensive applications in AI, cloud, enterprise and client environments.

The updated specifications include:

• NVMe 2.4 Base Specification
• Command Set Specifications (NVM Command Set 1.3, ZNS Command Set 1.5, Key Value Command Set 1.4, Subsystem Local Memory Command Set 1.3, Computational Programs Command Set 1.3)
• Transport Specifications (NVMe over PCIe Transport 1.4, NVMe over RDMA Transport 1.3, NVMe over TCP Transport 1.3)
• NVM Express Management Interface (NVMe-MI) 2.2 Specification
• NVMe Boot 1.4 Specification

Montage Technology today announced the industry's first trial production of its CXL 3.2 Memory eXpander Controller (MXC) chip, designed to address escalating demands for memory expansion and resource pooling in AI, cloud computing, and data center environments. Leveraging CXL technology, the MXC chip enables customers to build memory infrastructure with higher capacity, improved resource utilization, and greater flexibility.

The CXL 3.2 MXC complies with CXL Type 3 specifications, including both CXL.mem and CXL.io protocols. It is designed with PCIe 6.x and CXL 3.2 protocol support, delivering data transfer rates of up to 64 GT/s. The chip integrates dual DDR5 memory controllers that support up to 8000 MT/s DDR5 memory. By converting host-side CXL memory requests into DDR commands in real time, it enables efficient data exchange between the host and backend memory, effectively breaking through the memory capacity limits of traditional servers.

Microsoft has launched Xbox Backward Compatibility on PC, starting with four original Xbox titles: BLiNX: The Time Sweeper, Conker: Live and Reloaded, Crimson Skies: High Road to Revenge, and Fuzion Frenzy. All four games are available through the Xbox app on PC and included with every Xbox Game Pass tier. Digital console license holders can also access them via Xbox Play Anywhere. The feature requires Windows 11 and works on PC and supported handhelds including the ROG Xbox Ally and Xbox Ally X. Hardware requirements are modest, all you need is an NVIDIA GTX 950, AMD Radeon RX 550, or Intel UHD 770 / Arc A310 on the GPU side, a quad-core CPU, and 8 GB of RAM.

The company added several PC-specific enhancements like 4x resolution upscaling, V-Sync, fullscreen and windowed modes, anisotropic filtering, improved antialiasing, and adjustable language and audio settings. According to the official blog post published today by Jason Ronald, VP of Next Generation, Xbox, more backward-compatible titles are planned for later, however, no timing or hints about future titles have been made.

The Video Electronics Standards Association (VESA) today announced DisplayHDR True Black 1400, a new flagship performance tier in its DisplayHDR True Black specification that certifies the latest generation of OLED displays for professional HDR content creation and premium HDR content consumption. The new tier recognizes advances in OLED display technology that enable significantly higher luminance performance while preserving the deep black levels, exceptional contrast and visual fidelity that define the DisplayHDR True Black program. The first products certified to the DisplayHDR True Black 1400 specification will be showcased at Bilibili World 2026 in Shanghai, China, July 10-12.

DisplayHDR True Black 1400 is VESA's highest-performance DisplayHDR True Black tier to date, requiring certified displays to achieve at least 1400 cd/m² peak luminance while maintaining black levels as low as 0.0005 cd/m². The new tier also raises full-screen luminance performance to at least 700 cd/m², representing a significant advancement in sustained HDR brightness while maintaining the exceptional black-level performance associated with emissive display technologies such as OLED.

With the Steam Controller and the Steam Machine now officially launched and presumably making their way into the hands of consumers, the only new hardware product left for Valve to launch is the Steam Frame. Now, it seems as though the launch date for the new VR headset may be approaching, with Valve's Portal 2 getting a new update on June 29 that includes Steam Frame compatibility information. The official Steam Frame rating for Portal 2 is "Playable," with the game missing out on Frame Verified status due to not rendering in the Frame's native resolution. Curiously, Half-Life: Alyx, Valve's VR-only Half-Life game, has not yet received a similar compatibility update at the time of writing.

The Frame Verified rating has not yet been added to Steam's storefront, with compatibility information listed on SteamDB, but a June 29 update also fixed an issue with Portal 2 launching on Linux, suggesting Valve is putting in active development work to get its games and verification system ready for the official launch of the Frame hardware. According to the recently published Steam Frame Verified specifications, games need to be fully playable on the Frame controllers and reach at least 90 FPS for standalone VR titles and 30 FPS for standalone 2D titles at 1280 × 720 p, and it must have a legible UI at the default resolution of the Frame.

Yesterday, Qualcomm announced its High Bandwidth Compute (HBC) product, a memory-compute solution hybrid designed to replace traditional High Bandwidth Memory (HBM) and offer greater performance, efficiency, and throughput. HBC utilizes LPDDR memory, stacking them across multiple layers in a 3D vertical space and interconnecting them with through-silicon vias (TSVs). This approach provides better energy efficiency than traditional HBM, which stacks regular DDR in vertical layers, as the low-power DDR (LPDDR) chips consume less power while delivering similar bandwidth and capacity. Qualcomm has successfully stacked multiple LPDDR layers on top of each other, with the base of the HBC being a compute die that performs near-memory computation and offloads some processing from the main processor.

This technique is similar to what HBM4 memory currently does, with the base die being a logic die to better integrate compute solutions underneath for packet tracing and data preparation for entering and exiting HBM. However, Qualcomm aims to achieve higher efficiency and more optimized data movement, with a 6x increase in bandwidth per watt compared to the current HBM specification, likely HBM4. HBC Gen 1 achieved 133 TB/s of bandwidth on the AI250 accelerator card, a 18x increase over the LPDDR5X currently used in the AI200 card from Qualcomm. The company also claims there is a viable roadmap for using HBC across many Qualcomm AI accelerators, with HBC Gen 1 shipping with the AI250 AI accelerator in mid-2027. The next-generation HBC Gen 2 is expected to yield further significant gains in bandwidth.

Alongside a handful of new notebooks and pre-built desktop PCs, Colorful, the Chinese electronics brand widely known for its motherboards and NVIDIA GPUs, has revealed a chunky, white and gold iGame GeForce RTX 5090D v2 Arcanix 24 GB graphics card at Computex. The new GPU bears the brand's 30th anniversary branding and a bold white and gold color scheme, with what appears to be a massive 3.5- or four-slot cooler that has four fans in total—three on the front, with an extra fan towards the rear of the backplate over a flow-through section of the heat sink.

Colorful has not shared much about the GPU's specifications, but it will be based on the 5090D V2 GPU, and the show unit TechPowerUp's news team checked out at Computex has three DisplayPort ports and an HDMI port as outputs. There is also an easily accessible toggle switch on the I/O section labelled "Normal" and "Turbo," indicating that there will be a factory overclock available. It's unclear whether this will be a global launch for Colorful or simply a China-exclusive, like many RTX 5090D V2 models are. Currently, Colorful's flagship 5090D V2 GPU

Intel originally introduced the ATX12VO (ATX 12 Volt Only) standard in 2020, promising simplified PSU designs and efficiency improvements over the standard ATX standard, with the ATX12VO V2 standard arriving in 2022 with further efficiency improvements, especially when it comes to idle power draw. According to a recent leak by @momomo_us on X, ATX12VO appears to be getting yet another revision with further improvements to efficiency. It's not clear when Intel will officially announce the standard update, but the proximity to Computex suggests that a Computex 2026 launch may be possible.

The key takeaways for ATX12VO V3 appear to be that the update will introduce a new 8-pin connector, with the new standard eliminating the standby rail in the name of efficiency. Intel will also introduce new Low Power and High Power modes, which is meant to improve safety and efficiency. According to benchmark results from the leaked presentation slides, ATX12VO V3 is up to 29% more efficient than regular multi-rail ATX PSUs at idle and as much as 12% more efficient during load benchmarks. The new PSU standard also brings over PMBUS from the server world, which adds four optional motherboard connector pins, for a total of 12, and allows PSU-to-system communication for more advanced digital power management.

1047 Games, the studio behind Splitgate and Splitgate: Arena Reloaded, recently revealed that it was working on a new movement shooter inspired by Titanfall and Call of Duty Black Ops 3, seemingly in an attempt to not have all of its eggs in one basket. Not much has been officially revealed about the game via official or unofficial channels, but a Steam page for Empulse (stylized "EMPULSE"), and it turned out to be that very same game revealed previously by 1047 Games CEO, Ian Proulx. Despite the studio already previously accepting sign-ups for playtesting, there is no launch date or even launch window attached to the Empulse just yet, but the Steam Store page does divulge a handful of details and screenshots. The game has also been confirmed to be launching on both PS5 and Xbox consoles.

The Steam page confirms that Empulse will feature 6v6 gameplay with a focus on mobility, verticality, and a mix of gunplay and ability-based combat. The game will feature player-controllable mechs that spawn randomly across the map, and the map will be scattered with "Holojumps" that extend jump flight time and unlock parkour routes that would otherwise be unavailable to players. It will also feature grappling hooks and P.A.I.N.T bombs that alter surfaces in an as-yet unspecified manner. There is also a table of minimum system requirements that seems quite forgiving. If the system requirements are accurate, the minimum system specifications are an Intel Core i3-6100 or AMD Ryzen 3 1200, 8 GB of RAM, and an NVIDIA GeForce GTX 960 or AMD Radeon RX 470. The recommended specifications bump things up to an Intel Core i5-6600K or AMD Ryzen 5 1400, 12 GB of RAM, and an NVIDIA GeForce GTX 1060 or AMD Radeon RX 580 GPU.

Despite recent leaks of both Subnautica 2 and Forza Horizon 6, little has been said online about either game running on a Valve Steam Deck, even though Forza's Playground Games promised Steam Deck and ASUS ROG handheld compatibility when it revealed the minimum hardware requirements for the game. Now, on the eve of Subnautica 2's May 14 Early Access launch and the review embargo date for Forza Horizon 6, both games have been officially Steam Deck verified by Valve. While there are still questions about the performance of both games, it's yet another sign that two popular games from sizeable studios are being optimized for Valve's hardware—a positive indicator for the Steam Machine's playable library, too.

This means that both games have been tested for performance, UI scaling on the small, low-resolution display of the Steam Deck, and compatibility with both the Deck's controller and the Proton compatibility layer that SteamOS uses to run Windows games on Linux. It's unclear what sort of performance or resolution we can expect from either game on the Steam Deck, but Subnautica 2 officially calls for a minimum of an NVIDIA GeForce GTX 1660 or AMD Radeon RX 5500 XT and an AMD Ryzen 5 2600 or Intel Core i5-8400 CPU—those minimum specs will only grant you 30 FPS at 1080p and low quality, however, so Steam Deck gamers will seemingly be relying on tech like FSR to get reasonable frame rates. The same questions surround the handheld frame rates of Forza Horizon 6, since the published minimum hardware specifications make no mention of what quality and frame rates will be possible on the Steam Deck or ASUS ROG consoles.

ASUS's latest NVIDIA GeForce RTX GPU has just launched into its creator- and aesthetics-focused ProArt series as the ProArt RTX 5090 OC Edition, which ASUS says is designed for small form factor systems. The ASUS ProArt GeForce RTX 5090 OC Edition features the typical, clean, black and gold design with a 2.5-slot shroud and dual-fan design, and it features a "wood patterned laminate trim" along the curved edge facing the exterior of the PC case. The GPU is cooled by a vapor chamber with two heat sinks for flow-through cooling with hot air exhausting via vents in both the side and back of the GPU. The GPU itself is based on the NVIDIA Founder's Edition PCB, meaning uses the same angled 12V-2x6 power connector as NVIDIA's design and complies with the GeForce SFF-Ready specification. US pricing has not yet been revealed, but Geizhals has the ProArt RTX 5090 OC Edition listed at €3,505.89 ($4,115.21 converted).

ASUS claims that the cooler's design improves cooling efficiency by 11% and reduces the overall size of the card by as much as 27% compared to similar cards without the double flow-through design. It is also cooled by liquid metal, and the PCB is protected by a conformal coating, which helps in the event of leaking liquid metal and protects against high humidity environments. According to the GPU's spec sheet, the ProArt RTX 5090 OC Edition features a single HDMI 2.1b port, dual DisplayPort 2.1b ports, and a USB Type-C port, for a total of four displays. The GPU itself is clocked at 2482 MHz, with an OC Mode that boosts that to 2512 MHz. There's also a hardware switch on the GPU shroud that switches between Performance and Quiet modes—both modes feature the same power and frequency targets, but Quiet mode has a less aggressive fan curve, to keep noise down. Obviously, it is also compatible with ASUS GPU Tweak III, which allows users to tweak their own fan curves, power targets, and frequency in a GUI.

It seems as though AMD's Radeon RX 9000-series GPU family may finally be getting a true entry-level variant. According to a recent report by VideoCardz citing a source inside one of AMD's AIB partners, AMD will soon announce the AMD Radeon RX 9050 with 8 GB of GDDR6 VRAM and a down-clocked version of AMD's Navi 44 GPU also used in the RX 9060. AMD has not yet revealed any details about the potential upcoming GPU, but there is a spec sheet divulging some of the details of the GPU, including a higher core count than the RX 9060. The card is expected to require a minimum 450 W PSU, suggesting reasonably low power draw, although likely not low enough to power off the PCIe slot alone.

The aforementioned spec sheet lists a Game clock of 1920 MHz and a Boost clock of up to 2600 MHz for the RX 9050, although, notably, the actual stream processor count is up from 1792 on the RX 9060 to 2048 on the 9050. It's entirely possible that we will see the GPU modding community come up with some way to hack a higher clock speed onto the 9050 via something like a 9060 XT vBIOS or some similar trick, but until then it's entirely likely that performance will be lower than the 9060, because the clock speeds are slower on the 9050—13% lower Boost clocks and 20% slower Game clocks. The reports also note that the specifications are preliminary, so it may be the case that some of the stream processors will be disabled to put the 9050 below the 9060. This non-finalized spec means we are also unlikely to see the Radeon RX 9050 launch anytime soon. The spec sheet also lists a PCIe 5 ×16 bus width and an HDMI 2.1b and dual DisplayPort 2.1a ports.

PCI-SIG has released a small update on its upcoming PCIe 8.0 standard, with the draft milestone reaching version 0.5. Perhaps the most intriguing aspect of this draft update is not the performance itself, but the exploration of a new connector technology to support this high-bandwidth protocol. Last year, we learned that PCI-SIG plans to implement a 256.0 GT/s raw bit rate and 1 TB/s of bidirectional bandwidth in the x16 lane configuration. We had assumed that the protocol would continue using the familiar connector technology seen in previous PCIe updates. However, it turns out that the current connector might be a limiting factor, prompting the search for a replacement for the traditional PCIe electrical connection.

The traditional PCIe connector is a copper-based link with up to 16 lanes connecting graphics cards to a slot. In a full x16 lane configuration, the PCIe generation supported by the motherboard provides the best performance, offering the maximum bandwidth the platform can deliver. However, with a 256 GT/s raw bit rate, the connector provides about 1 TB/s of bidirectional bandwidth, which is eight times faster than the current PCIe 5.0 platform used with modern GPUs and CPUs. This indicates that the current physical layer facilitating communication between a GPU and a motherboard is nearing saturation with the advent of PCIe 8.0, necessitating the consideration of an alternative connection method.