AMD has begun integrating its next-generation graphics IP into the Linux kernel, releasing a series of eight patches that add GFX13 support to the open-source AMDGPU driver. These patches, submitted to the mailing list on November 9 by AMDGPU maintainer Alex Deucher, contain over 12,600 lines of code, excluding the typically large header files. GFX13 is anticipated to be the graphics core for RDNA 5, although AMD has not officially confirmed this connection. However, since GFX12 has been linked to RDNA 4, it is assumed that GFX13 is the correct codename for RDNA 5.

The patches introduce version 13 of three IP blocks: the "graphics" GFX block itself, the MicroEngine Scheduler (MES), and the IMU. On the compute side, the AMDKFD driver gains initial GC 13 support, a CWSR trap handler for saving and restoring compute wave state, and a temporary change that disables queue reset in the topology. The final patch enables the MGCG, CGCG, and CGLS clock gating features, which reduce power consumption when parts of the GPU are idle. Interestingly, the series mentions two specific targets, GFX 13.0.1 and GFX 13.0.4, indicating that AMD is preparing at least two graphics IP configurations. These version numbers do not correspond to any specific GPU models, so it is unclear which products they relate to. RDNA 5 is expected to be a significant advancement over RDNA 4, necessitating numerous patches to add support.

Police in Oakland, California, are investigating the theft of six pallets of NVIDIA hardware worth $11.4 million that were bound for Fort Worth, Texas, the Bay Area News Group reports. A truck carrying a food transportation company's logo picked up the cargo in Oakland on August 29 as planned, but the truck and its drivers then went off the grid heading south. According to authorities, the truck made several multi-hour stops before police lost track of it in San Bernardino County. Apparently, nobody realized the cargo was gone until early September, and Oakland police say the theft was reported to them on September 13. No arrests have been announced.

What exactly was on those pallets hasn't been disclosed, so we don't know if it was AI accelerators, complete systems, or GeForce graphics cards. Speaking to ABC7, UC Berkeley professor Olaf Groth said NVIDIA chips can sell for roughly double their US price abroad, with a $550,000 compute system going for $1 million or more. He added that the heists are getting bigger and more complex, which points to increasingly professional crews.

When ASRock joined the group of AIBs patching Radeon RX 7900 XT/XTX GPUs late last month, everything seemed fine for a few days. However, multiple users are now reporting that their ASRock GPUs have been bricked after updating the firmware with the latest VBIOS. ASRock has released VBIOS updates for ten of its Radeon RX 7000 series graphics cards, joining Sapphire and GIGABYTE. The new firmware, version 1.0.5, was made available on ASRock's support pages on October 29 and includes updates for the RX 7600 XT Challenger and Steel Legend, RX 7700 XT Challenger, RX 7800 XT Challenger and Steel Legend, RX 7900 XT Phantom Gaming (including the White version), and the RX 7900 XTX Phantom Gaming and Taichi, along with the Taichi White.

However, several users have reported that VBIOS version 1.0.5 bricked their cards. The issue occurs when users start the ASRVBFlash 1.0.5 utility; Windows begins its reboot sequence, and the system then fails to boot. A solution to this problem appears to be reinstalling an older VBIOS, which some users have downloaded from TechPowerUp's VBIOS hosting. This can help if your PC is bricked by any VBIOS update, and you need to find an older version to flash.

NVIDIA's GeForce RTX 50-Series "Blackwell" gaming graphics cards are expected to see significant price adjustments this week. According to Chinese ChannelGate, add-in card (AIC) partners have confirmed that price hikes are scheduled due to the ongoing demand for these GPUs outstripping supply. Some major AIC partners, like Colorful and Galax, have reportedly already started sharing information about these price increases. The most significant example is expected to be for SKUs based on NVIDIA's GeForce RTX 5090 DV2, which will see an increase of up to 3,300 RMB, translating to nearly $500. Even entry-level GPUs are not exempt, as the GeForce RTX 5060 is expected to see around a $30 price increase on existing models. These changes are anticipated to be implemented this week for retail partners, meaning that consumer prices might take a week or two to reflect the new listing prices.Below is the list of price hikes in China, expected to be the norm.

ASRock has posted vBIOS updates for ten of its Radeon RX 7000 series graphics cards, joining Sapphire and GIGABYTE. The new firmware, version 1.0.5, went up on ASRock's support pages on October 29 and covers the RX 7600 XT Challenger and Steel Legend, RX 7700 XT Challenger, RX 7800 XT Challenger and Steel Legend, RX 7900 XT Phantom Gaming (including the White version), and the RX 7900 XTX Phantom Gaming and Taichi, along with the Taichi White. The files are flashed through ASRock's ASRVBFlash tool on Windows. Owners of these ASRock cards are advised to check the vBIOS section of their specific model's support page before flashing.

ASRock doesn't say what the update actually changes, as the description field only lists the flash utility version. However, the timing points to the same RDNA 3 security fix that Sapphire began offering in early October. GIGABYTE later tied its own files to the AMD-SN-6023 identifier, and AMD confirmed the updates address potential security vulnerabilities, adding that it has no reports of them being exploited. AMD still hasn't published the bulletin itself, so the details remain unknown. Interestingly, ASRock's own download page still carries its standard warning that it does not recommend updating the vBIOS if the card is already running normally, which reads a little odd next to AMD's advice to update as soon as firmware is available.

AMD's Radeon RX 9070 GRE graphics card, initially designed for customers in China and later sold in more markets, has officially reached a frequency of 3.9 GHz. TechPowerUp Forum user "ShrimpBrime" created custom tools for Radeon GPUs called "R-Team OC" and "GPU Frequency Generator." Using the stock XFX Radeon RX 9070 GRE Swift GPU without any VBIOS modifications, specialized cooling, or other hardware changes, the overclock was stable for some time. However, the driver crashed and recovered shortly after the screenshots were taken. Utilizing the custom-developed RTOC tool for Radeon tuning and monitoring, the GPU was loaded with regular triangles in DirectX 9 instead of a demanding 3D title, meaning that this OC is not applicable in real-world testing, but serves as a proof of concept.

Using a customized GPU frequency generator built in DirectX 9, the GPU rendered a simple scene containing 2,048 triangles per draw call and 2,048 draw calls per frame, totaling 4,194,304 triangles. This is light work for modern graphics cards but just enough to maintain a high frequency. The developer says the 3.9 GHz reading was maintained under load rather than appearing as a momentary telemetry spike, while approximately 3.85 GHz was reproduced repeatedly, including with AMD's own Radeon Software overclocking controls. To verify the readings, ShrimpBrime compared GPU-Z and HWiNFO64 with telemetry gathered through AMD's ADL and ADLX interfaces. RTOC polls shared monitoring data every 2 ms, although fresh power-management telemetry reportedly arrives roughly every 15.5 ms.

Palit issued a statement today to address and dismiss current industry rumors that GALAX GPU factories are winding down. The rumor suggested that GALAX and its European-focused KFA2 brand were experiencing supply issues with NVIDIA. However, Palit has clarified that these reports are inaccurate. As one of NVIDIA's largest add-in card partners, Palit sells a significant volume through its brands. The speculation that GALAX factories are shutting down due to supply problems with NVIDIA would imply a risk to NVIDIA's entire business with its AIC partners, which doesn't make much sense. Palit explained that since Palit and GALAX are now operating under one roof, they share some manufacturing facilities, and GALAX products continue to be sold.

Palit has also confirmed that the company is developing next-generation GALAX Hall-of-Fame (HOF) graphics cards, along with regular GALAX and KFA2-branded GPU models. While the current generation of GPUs remains available, the mention of next-generation GPUs in development is particularly intriguing. This indicates that the upcoming NVIDIA GeForce RTX 60-Series "Rubin" gaming GPUs will also be available in GALAX HOF and KFA2 variants, providing extreme overclockers with more HOF generations in the future.

TSMC is reportedly considering raising prices for its CoWoS (Chip on Wafer on Substrate) advanced packaging by as much as 20%, according to supply chain sources cited by Taiwan's Commercial Times. Foundries and outsourced packaging firms are apparently weighing another round of 10% to 20% increases, as CoWoS demand has spread beyond AI GPUs to server CPUs and custom ASICs, leaving a growing backlog of unfilled orders. TSMC hasn't confirmed anything yet.

Testing is turning into a bottleneck of its own. Nomura estimates, via Seoul Economic Daily, that NVIDIA's Blackwell takes roughly four times longer than Hopper in final testing, while Rubin could need seven times as long. Packages keep growing too, with TSMC already mass producing CoWoS-L at 5.5 times the reticle limit. Tool lead times have reportedly stretched from three to six months to as long as eight months. TSMC, which usually negotiates for discounts or extra features, apparently placed this year's equipment orders at existing prices instead, which says a lot about how tight things are.

MSI proudly announces the launch of the limited MSI GeForce RTX 5080 16G COD MW4 EDITION OC Graphics Card, inspired by the intense action of Call of Duty: Modern Warfare 4. Developed in close collaboration with Activision, this exclusive edition brings together MSI's cutting-edge graphics technology and the unmistakable identity of the Call of Duty franchise.

Featuring a distinctive, custom-designed aesthetic created exclusively for this collaboration, the graphics card captures the spirit of Call of Duty: Modern Warfare 4 through a bold, military-inspired design that sets it apart from standard graphics cards. Combining high-performance gaming technology with a one-of-a-kind design, this limited edition is a collector-worthy centerpiece for gamers and Call of Duty fans.

AMD has finally commented on the vBIOS updates that Sapphire and Gigabyte are pushing out for Radeon RX 7000 series graphics cards, though the statement doesn't clear up much. The company reached ComputerBase through Sapphire, which the German outlet had asked last week about what the new firmware actually fixes. Sapphire started offering the updates earlier this month for its RX 7600 through RX 7900 XTX models, citing only a security issue identified in partnership with AMD. Gigabyte followed and was a bit more specific, pointing to vulnerabilities in the AMD VBIOS under the AMD-SN-6023 identifier, an entry that still doesn't show up on AMD's product security page. Here's what AMD said:

As part of our standard procedure, AMD is releasing a firmware update to address potential security vulnerabilities. These fixes are provided ahead of public disclosure to give customers time to update their firmware. As a best practice, we recommend users update the firmware of their graphics cards as soon as it is made available by their respective GPU manufacturer. There are no reports of these vulnerabilities being exploited in practice.

AMD's next-generation gaming graphics cards featuring the RDNA 5 GPU architecture are unlikely to debut before the second half of 2027, according to well-known hardware leaker Kepler_L2. The leaker suggests that AMD is unlikely to release any RDNA 5 desktop graphics cards before Computex 2027, indicating that these GPUs will probably arrive in the latter half of that year. AMD is expected to lead with the silicon codenamed "AT2," which is part of a family of discrete RDNA 5 GPUs. This will be followed by "AT0," "AT3," and "AT4," which are expected to launch a few quarters later in 2028. Since "AT2" likely refers to a mid-range design, AMD might begin its rollout with a mid-range GPU first, followed by flagship and lower-end models in 2028.

Previously, we learned that AMD's RDNA 5 GPUs will incorporate GDDR7 memory with unknown capacities, as GDDR7 is now listed among the supported VRAM types in the AMDGPU driver. Currently, AMD's graphics cards use GDDR6 for the consumer RDNA 4-based Radeon RX 9000 series, while Instinct MI data center and workstation GPUs use HBM3E and HBM4 memory. Since AMD is developing data center GPUs with HBM and consumer cards are the only models using GDDR, the GDDR7 placeholder suggests future support for RDNA 5.

Vince "KINGPIN" Lucido has blamed NVIDIA for the troubled launch of the EVGA GeForce RTX 3090 Ti KINGPIN, the last graphics card to carry his name. In the first video on his new Overclocking Kingpin YouTube channel, the overclocker said he wanted the card to have two 12VHPWR power connectors, while NVIDIA wanted it limited to just one. The disagreement, in his words, "really screwed up the production timeline", and he also holds it responsible for how few units were made afterward. The card did eventually ship with both 16-pin connectors when it went on sale for $2,500 with a bundled 1600 W power supply, about three months after the RTX 3090 Ti itself came out. This is only Lucido's side of the story, and it isn't clear why NVIDIA pushed for a single connector.

In September 2022, EVGA ended its partnership with NVIDIA and stopped making graphics cards, with CEO Andrew Han pointing to a souring relationship between the two companies. Every card in development was canceled, KINGPIN models included. Lucido is now working with ASUS, which sent over the hardware shown in the video, although no ASUS-made KINGPIN card has been confirmed. He did lay out one condition for such a card: the PCB and power delivery have to let the GPU reach its full potential under liquid nitrogen without the buyer modifying anything. An earlier plan with PNY never produced a card, and Lucido ended up skipping the RTX 50 generation entirely.

AMD has confirmed that its FSR 4 machine learning upscaling technology will be available for APUs with integrated graphics by the end of this year. Jack Huynh, senior vice president and general manager of the computing and graphics group, shared this news with the South Korean media in Seoul. He mentioned that FSR 4 will be coming to more devices as AMD optimizes the software stack for APUs. Initially, FSR 4 was only available on RDNA 4-based Radeon RX 9000 series graphics cards. This summer, AMD added support for FSR 4 on RDNA 3 hardware with Radeon RX 7000 series discrete graphics cards. However, APUs like "Strix Halo," which feature RDNA 3.5 integrated graphics, were not supported. Now, it appears these models will receive official support by the end of the year. Below is the machine-translated official quote:

AMD's Jack HuynhWe will expand FSR 4 across our product lineup, including APUs, gaming laptops and handheld devices, by the end of this year.

NVIDIA today introduced its GeForce 617.42 WHQL Game Ready driver, bringing support and optimizations for various new game releases. First, there is Game Ready support for two new titles: Call of Duty: Modern Warfare 4 and STAR WARS: Galactic Racer. These two AAA games are newly supported with this driver version, meaning the experience should be ready to go, but keep an eye out for any possible issues, as with every newly supported game. Next, we have optimizations for Dragon's Dogma 2: Dark Arisen and Valor Mortis, which have been fine-tuned to work better on NVIDIA GeForce RTX graphics cards, extracting more performance and reducing visual glitches. Support for updates like Microsoft Flight Simulator 2024 Sim Update 7 has also been included, ensuring the package covers modern game updates and releases.

The driver itself also fixes an issue with Assassin's Creed Shadows, which previously crashed to the desktop after extended gameplay with driver version 616.56. Gamers running this specific version should upgrade to the latest 617.42 WHQL edition, and the full changelog can be found below.

DOWNLOAD: NVIDIA GeForce 617.42 WHQL

A modder going by SupraGSX has released an unofficial Windows XP driver on GitHub that adds support for NVIDIA's GeForce GTX 10-series "Pascal" graphics cards, including the TITAN Xp. Called Forceware 382.69, it is a patched version of NVIDIA's 368.81 driver from July 2016 and only works on 32-bit Windows XP. The timing is a bit funny, since NVIDIA ended regular driver support for Pascal last year and later dropped the cards from its 590 driver branch. However, Forceware's latest release covers the GTX 1060, GTX 1070, GTX 1070 Ti, GTX 1080, GTX 1080 Ti, TITAN X (Pascal) and TITAN Xp, along with some GTX 900-series and Quadro cards.

It also adds DisplayPort 1.4 HBR3 and HDMI 2.0 handling. SupraGSX notes that these are INF additions, and that not every model was individually validated. The October 4 build fixes OpenGL crashes on Pascal, which was verified on a GTX 1080 Ti. Support for GP107 cards such as the GTX 1050 Ti is experimental and untested on real hardware. X user @TheBobPony, who shared a GPU-Z screenshot of a GTX 1080 running on Windows XP, described the driver as "vibe coded." That claim remains unconfirmed currently, as the project's README and release notes make no mention of AI tools. Readers are cautioned to keep unverified drivers off their primary PCs, AI-coded or otherwise. Still, the effort deserves some credit, if only for keeping the Xp on XP meme alive.

After EVGA's abrupt exit from the GPU space, high-profile Vince Lucido, aka Kingpin, seemingly took a hiatus from the PC overclocking scene, aside from briefly working with PNY on a planned RTX 50-series launch before scrapping the project. Now, as per a recent video update on his YouTube channel, Lucido has found another hardware partner for his signature overclocking GPUs in ASUS.

In the update, Lucido confirms that he has "some really sick hardware from ASUS and Corsair to start with," and that he has had "numerous talks" with high-level management at ASUS, which are what ultimately persuaded him to get back into the overclocking industry. He has promised that whatever he has planned for PC overclocking will appear on the Overclock Kingpin YouTube channel, which has no published videos at the time of writing.

Microsoft's Advanced Shader Delivery (ASD), which provides pre-compiled shaders with the game itself, is finally becoming available on a range of platforms. The company confirmed that this feature will expand to NVIDIA's GeForce RTX graphics cards and the RTX Spark platform later this month. This means that game developers will be able to ship pre-compiled shaders to speed up game loading times and reduce any shader-induced stuttering during gameplay. When Microsoft unveiled a public preview of ASD, it reportedly supported various AMD RDNA generations, but now more platforms are receiving support.

Currently, ASD is claimed to work on AMD RDNA 1, RDNA 2, RDNA 3, RDNA 3.5, and RDNA 4 GPU generations, as well as on Qualcomm Snapdragon X2 Series with Adreno integrated graphics. Later this month, Microsoft plans to extend ASD to Intel Arc B-Series, which includes B570, B580, Intel Core Ultra Series 2, and Intel Core Ultra Series 3. This update will coincide with support for all NVIDIA GeForce RTX GPUs and the RTX Spark platform, ensuring that every RTX-based GeForce system benefits from the speedup.

Intel today updated its Compute Runtime with support for Portable Instruction Set Architecture (PISA) format, a new virtual ISA and programming model for Intel GPUs. Unlike CPUs, where software commonly targets established ISAs such as Arm or x86, GPUs have their own native instruction sets and hardware-specific architectures. As GPU generations introduce new hardware capabilities, execution features, and native instructions, compilers targeting the machine ISA have to account for differences between architectures. A standardized virtual ISA changes this by introducing an abstraction layer between compiler output and the GPU's native machine code. Essentially, PISA provides a common compiler target that allows generalized code to be written once and then finalized for multiple compatible Intel GPU architectures.

With PISA, Intel aims to provide a stable and portable ISA for compilers that target Intel GPUs. This means that a workload compiled to PISA can potentially run across multiple compatible Intel GPU architectures without requiring architecture-specific code generation for each generation. The PISA representation can then be finalized into native machine instructions for the target GPU, allowing the same workload to take advantage of the capabilities of newer hardware where supported, without requiring changes to the original PISA code. This does not mean that every PISA workload will automatically run on every Intel GPU, however, as compatibility depends on the target architecture and the features and instructions used by the program. PISA is also not Intel's first virtual GPU ISA, as the company's existing vISA has long served as an intermediate representation for its GPU compiler stack, while PISA introduces a newer portable ISA and programming model.

Sapphire is preparing a Valor Mortis edition of its Nitro+ Radeon RX 9070 XT OC, a themed variant created in collaboration with dev studio One More Level and publisher Lyrical Games to coincide with the launch of their first-person Soulslike action game on October 13, 2026. This isn't Sapphire's first themed graphic card, the most recent was the Nitro+ RX 9070 XT OC Crimson Desert edition. The Valor Mortis variant follows the same path: a custom logo on the fans and a magnetic backplate decorated with game artwork. Even if not officially announced yet, the GPU, memory, cooling, and factory overclock likely won't change as the specs would match the Crimson Desert edition. This means the GPU will have up to 3060 MHz boost clock and 2520 MHz game clock, while the card will feature 16 GB of 256-bit GDDR6 at 20 Gbps.

A copy of the game comes bundled with the card, however, Sapphire did not mention which edition, standard or deluxe. Also, pricing or an exact release date has not been announced. We estimate that the Sapphire Nitro+ Radeon RX 9070 XT OC will be available around October 13 or shortly after, given the game's launch date. As a side note, Sapphire is celebrating its 25th anniversary this year. Founder and CEO KD Au said the milestone "belongs to our customers, our partners at AMD, and every member of the SAPPHIRE team," adding that the company started with the ambition of building the best graphics cards in the world, a goal he says still drives the company today.

Colorful Technology Company Limited, a leading brand in gaming PC components, gaming laptops, and Hi-fi audio products, is opening a new chapter in the Japanese market through a strategic partnership with CFD Sales Inc. (CFD), a well-established Japanese hardware distributor. As part of this renewed market expansion, COLORFUL's flagship iGame graphics card series and BATTLE-AX motherboards will officially go on sale in Japan on September 16, ahead of TOKYO GAME SHOW 2026. The iGame graphics card series will also be showcased as part of CFD's new gaming brand, Crowxis, at the event.

Beginning September 16, Japanese customers will be able to purchase COLORFUL iGame graphics cards and BATTLE-AX motherboards through CFD's local distribution channels. Product and availability details will be available on the official CFD website. COLORFUL's partnership with CFD represents an important step in strengthening its presence among Japanese PC DIY enthusiasts and gamers, while bringing its latest graphics card and motherboard products to the Japanese market.

NVIDIA has launched the NBA 2K27 DLSS 5 Poster Challenge, giving away four custom RTX 5090 graphics cards as prizes. Given what the card actually costs to buy right now, the timing is a little funny. To enter, players capture an in-game NBA 2K27 screenshot or clip with DLSS 5 enabled on an RTX 50-series GPU, or through GeForce NOW Ultimate streaming from RTX 5080 servers, then post it tagging @NVIDIAGeForce with the hashtag RTXPoster. Easier entry paths exist too, like voting on other posters or replying to prompts on GeForce's social channels. The contest runs through December 31, with winners announced by the end of January 2027.

NVIDIA values each card at $2,599, including the custom NBA 2K27 wrap. That number looks almost quaint next to what the RTX 5090 actually sells for. Multiple retail trackers put US street prices well past $5,000 this month, more than double the card's $1,999 launch price, largely thanks to the GDDR7 memory crunch that has been squeezing the entire GPU lineup. Prices have kept climbing for months now, with no sign of letting up. So, oddly enough, winning a poster contest might be one of the more realistic ways to grab an RTX 5090 without draining your bank account. It's a fun bit of marketing for a card that's become something of a punchline for how far GPU pricing has drifted, though I doubt NVIDIA meant it that way.

A Chinese manufacturer is now selling a modified GeForce RTX 5090 with 96 GB of memory, three times the 32 GB on the retail card. Shenzhen Suqiao Intelligent Technology, an OEM/ODM with over a decade of experience building graphics cards, motherboards, and servers, listed the card on Alibaba for $3,888. That's roughly 35% less than a standard RTX 5090 currently costs in the US, and a fraction of what NVIDIA charges for its RTX PRO 6000 Blackwell. The latter uses the same GB202 die and now starts at a whopping $16,000.

Suqiao likely built this with a custom PCB running clamshell memory, packing GDDR7 modules on both sides of the board to double capacity, the same approach NVIDIA itself uses on the Pro 6000. The GPU die may come from harvested RTX 5090 units, reballed onto the new board and running on leaked or modified firmware, a method Chinese workshops already used to push the RTX 4090 to 48 GB VRAM. The listing does raise some doubts, though, since it mentions GDDR6X memory at 14 Gbps, which doesn't line up with a genuine 96 GB Blackwell build.

PC GPU shipments are actually increasing despite record-high prices, according to Jon Peddie Research. However, the details reveal more complexity. JPR measures shipments of all PC GPUs, and since PCs come in various forms, so do GPUs. Overall, in Q2 of this year, there was a significant 10.4% increase quarter-over-quarter and about 1.1% year-over-year compared to Q2 2025, reaching 75.5 million GPUs this quarter. This large increase is primarily driven by shipments in the notebook graphics card sector, which saw a massive 16.8% jump, while dedicated desktop GPUs experienced a 4% reduction on a quarterly basis.

Looking at the per-vendor statistics of PC GPUs, the three major GPU designers—AMD, Intel, and NVIDIA—show some changes compared to the same period a year ago, with slight adjustments to quarterly changes. Intel continues to dominate, with 56% of all PC GPU shipments being Intel-derived. Although the discrete desktop GPU segment might not be Intel's strongest area, it's important to remember that Intel sells millions of GPUs in laptop form factors and with integrated graphics across its SoCs. Most modern Intel CPUs include some form of integrated graphics. However, Intel's lead decreased by 1% quarter-over-quarter and about 5% year-over-year. NVIDIA saw a 0.46% increase, while AMD experienced a marginal 0.6% increase in market share quarterly.

A new tool called DLSS 5 Swapper, published on GitHub by developer rakanki911, automates installing DLSS 5's Neural Rendering into games, including ones that never officially got support for it. Users can drag and drop a game into the app, or let it scan the drive for installed titles on its own. For games that already ship with DLSS, the swap is simple. For everything else, the tool relies on DLSS5-Feeder, a community mod that emulates the DLSS calls required to trigger Neural Rendering, paired with ReShade for the depth buffer and motion data that DLSS needs to work. That combination reportedly runs across titles from DirectX 9 through DirectX 12, and even on a handful of emulators like PCSX2, Dolphin, and Xenia.

Perhaps even more interestingly, the app also extends DLSS 5 support to RTX 20-series and RTX 30-series graphics cards, sidestepping NVIDIA's RTX 50-series requirement. This tool builds on the same leaked DLSS 5 files that surfaced in NBA 2K27's pre-launch build, files modders have spent weeks getting to run on older GPUs and non-DLSS games. DLSS 5 Swapper mostly bundles that scattered work into a single app; however, since it's unofficial and has no ties to NVIDIA, it's important to verify the download source before installing anything. Given how far these mods have already spread, this tool is likely to become the most popular way for people to try Neural Rendering, at least until NVIDIA offers something similar itself.

The TechPowerUp team visited the Chieftec booth at IFA 2026, with some interesting offerings on display. Chieftec brought its case lineup to the show, mixing an unannounced E-ATX tower with two chassis it already launched earlier this year. The new one is the Atlas, an E-ATX case supporting up to 420 mm radiators in the front or top. It has room for three 120 mm fans behind the power supply shroud, one rear fan, and an extra 120 mm unit near the PCIe brackets meant to pull hot air away from the graphics card. There are eight expansion slots and up to 420 mm of GPU clearance. This case is really meant for workstation or server builds, and its clean, functional design makes it ideal for that purpose.