Klevv has updated its DDR5 Gaming/OC memory lineup, introducing new features and aesthetic options. The series that are receiving these updates include the Fit V, Bolt V, CRAS V RGB, and Urbane V RGB.

Klevv has expanded the colour palettes of its Bolt V and Fit V memory series. The Bolt V series now has white sticks, and the Fit V series now features black models. In addition, Klevv has reduced latency across its CRAS V RGB, Urbane V RGB, Bolt V, and Fit V series. The 6000MT/s variants now achieve latencies as low as CL26 or CL28, while the 6400MT/s models reach CL30, depending on the specific model.

To ensure seamless compatibility and a consistent user experience, Klevv is introducing “Optimised for AMD” models (AMD EXPO) across its DDR5 Gaming/OC memory range. These models are labelled on the packaging, allowing users to identify compatible SKUs for their AMD-based systems easily.

KLEVV's updated DDR5 Gaming/OC memory module series is slated for release in Q2 2025. All these modules are backed by a limited lifetime warranty, providing users with peace of mind.

KitGuru says: Have you ever owned a Klevv memory kit? Did these updates convince you to get one if the price is right?

The post Klevv updates DDR5 memory lineup with lower latency profiles, AMD EXPO support and new colour options first appeared on KitGuru.

Leaked presentations have revealed AMD's next-generation mobile CPU, codenamed “Gorgon Point”, offering a glimpse into the company's roadmap beyond the current Ryzen AI 300 “Strix Point” architecture. This 2026 successor promises a slight improvement in CPU performance while mainly maintaining the core CPU, iGPU, and NPU architectures.

The presentation shared by Naver (via @harukze5719) shows that “Gorgon Point” is expected to integrate up to 12 CPU cores, using a combination of “Zen 5” and “Zen 5c” architectures, an iGPU based on the RDNA 3.5 graphics architecture, and an NPU powered by XDNA 2. While the core IP remains consistent with “Strix Point”, performance enhancements are anticipated across the CPU, iGPU, and NPU. However, don't expect anything game-changing, as the architectures are still the same.

Image credit: Naver (via VideoCardz)

AMD's presentation included internal performance benchmarks, showcasing marginal improvements in single- and multi-threaded CPU performance. This suggests that the core configuration, likely a 4x “Zen 5” + 8x “Zen 5c” arrangement, will be carried over. However, if the core configuration is the same, there must be something else that would result in this performance improvement, like an increase in CPU clock frequencies or a change to the design of the “Zen 5” P-cores (“Zen 5+”?).

Furthermore, two “Gorgon Point” SKUs (a Ryzen and a Ryzen 7) will feature an enhanced NPU, delivering more than 55 AI TOPS, an improvement of 5 AI TOPS over the Ryzen AI 300 series top performers. However, most SKUs seem to pack the same XDNA 2 NPU found in Strix Point.

KitGuru says: As it seems, Gorgon Point is to Strix Point what Hawk Point is to Phoenix Point. As such, if you're already on a Ryzen AI 300 series, there's likely no point in upgrading.

The post Leak claims AMD Gorgon Point chips will feature Zen 5 CPU cores and RDNA 3.5 iGPU first appeared on KitGuru.

Nvidia App's latest update (11.0.3.218) introduces some interesting new features, including an AI assistant named Project G-Assist. This update also introduces DLSS Super Resolution scaling controls and incorporates key settings from the legacy Nvidia Control Panel, aiming to consolidate functionality within the new application.

Nvidia's Project G-Assist provides practical in-game assistance, coaching, and support. Nvidia describes it as an AI driven by a specialised Small Language Model (SLM), trained to perform specific tasks. These include optimising in-game settings, providing hardware information, and controlling lighting and fan speeds on compatible peripherals. Some games will also feature custom support for an enhanced experience. G-Assist's functionality is restricted to users with Nvidia RTX 30, 40, or 50 series GPUs equipped with 12 GB or more of VRAM. The System Assistant and Voice Command modules require approximately 6.5 GB and 3 GB of free disk space, respectively.

In addition to the AI companion, the Nvidia App recently received a DLSS Override feature, allowing users to fine-tune DLSS Super Resolution scaling. Instead of relying on fixed presets, the new update enables users to adjust the input resolution between 33% (DLSS Ultra Performance) and 100% (DLAA), providing granular control for optimal image quality and performance. This is particularly useful for users seeking a middle ground between two presets.

Furthermore, the Nvidia App now integrates two essential features previously exclusive to the Nvidia Control Panel: display scaling and colour settings. The former enables hardware-accelerated integer scaling, overriding in-game and application scaling, while the latter allows users to adjust display colour profiles. The latest version of the Nvidia App is available for download from the official Nvidia website or via an in-app update.

KitGuru says: Have you already updated your Nvidia App?

The post Nvidia App update introduces G-Assist, DLSS Super Resolution custom scaling and more first appeared on KitGuru.

Samsung has unveiled its 2025 lineup of Odyssey gaming monitors and the ViewFinity S8. The Odyssey portfolio features the groundbreaking Odyssey 3D, the new 27- and 32-inch Odyssey OLED G8 displays, and the ultrawide Odyssey G9.

The new Odyssey 3D (G90XF model) 27″ display introduces a glasses-free 3D gaming experience. Advanced eye-tracking technology and a patented lenticular lens create a natural-looking high-definition 3D image accessible via the Reality Hub app. Samsung is collaborating with game developers to optimise the use of this 3D technology, including Nexon for The First Berserker: Khazan and Neowiz for Lies of P. More developers should be added in the future.

In addition to 3D gaming, the Odyssey 3D features AI video conversion, which can transform video footage into 3D. Every scene of compatible content is analysed and converted to 3D, delivering consistent 3D effects and reducing eye strain. The Odyssey 3D also boasts impressive gaming performance, packing a 4K panel with a 165 Hz refresh rate and 1 ms GtG response time. The new Samsung 3D monitor is Nvidia G-Sync compatible and supports AMD FreeSync Premium. Moreover, it supports Edge Lighting, which enhances 3D immersion by illuminating the gaming area, matching the screen's lighting (like Philips Ambilight).

Moving on to the non-3D gaming monitors, the new Odyssey OLED G8 (G81SF) is available in 27-inch and 32-inch variants. Both monitors are VESA DisplayHDR TrueBlack 400-certified, have glare-free technology certified by Underwriters Laboratories (UL), a 240 Hz refresh rate, and a G2G response time of 0.03 ms. Like the Odyssey 3D, the new G8 supports AMD FreeSync Premium Pro and is Nvidia G-Sync Compatible.

Samsung OLED Safeguard+ protects screens using its dynamic cooling system, which features the industry's first pulsating heat pipe. Samsung claims this system dissipates heat five times more effectively than traditional graphite sheets without affecting brightness, preventing burn-in and extending the screen's lifespan.

The Odyssey G9 (G91F model) makes ultrawide gaming more accessible by streamlining the experience. Its 49-inch Dual QHD display features a 1000R curve, a 144Hz refresh rate, and a 1 ms response time, providing responsive gameplay. AMD FreeSync Premium Pro eliminates tearing and stuttering, while the VESA DisplayHDR 600 certification suggests it can offer an acceptable HDR experience. The Odyssey G9 also offers impressive multitasking capabilities, including PbP and PiP modes. Moreover, the Odyssey G9's Auto Source Switch+ feature helps players quickly access games by detecting connected devices and automatically displaying them when turned on.

In addition to the 2025 gaming monitors, Samsung is introducing the 37-inch ViewFinity S8 (S80UD), the largest 16:9 4K Samsung monitor to date. The ViewFinity S8's 4K screen is approximately 34% larger than the previous model, maximising productivity and efficiency. The ViewFinity S8 is designed for comfort and productivity, featuring a 10-bit panel that displays 1 billion colours. TÜV Rheinland has certified its ergonomic design as an “ergonomic workspace display”, and Intelligent Eye Care is TÜV-accredited for reducing eye strain. A 90W USB-C port allows connecting three devices, and a KVM switch enables control of multiple inputs on the display, simplifying the integration of work and personal devices.

The new Odyssey 3D, OLED G8 and G9 are available for pre-order starting today.

Discuss on our Facebook page, HERE.

KitGuru says: Would you fancy yourself any of Samsung's new monitors?

The post Samsung showcases new 3D gaming display and 4K monitors first appeared on KitGuru.

Funcom has officially announced the release date for Dune: Awakening, the anticipated multiplayer survival game inspired by Frank Herbert's seminal sci-fi novel and Denis Villeneuve's acclaimed film adaptations. Players can brave the treacherous sands of Arrakis on May 20th, with pre-orders now live.

Three distinct editions are available for pre-order on Steam: the standard edition, the Deluxe Edition, and the Ultimate Edition. The latter includes a 90-minute digital soundtrack, for which Funcom has released a behind-the-scenes video showcasing composer Knut Avenstroup Haugen's recording session with a full orchestra at London's renowned AIR Lyndhurst studio.

The standard edition of Dune: Awakening is priced at £41.99. The Deluxe Edition, which is priced at £58.50, and Ultimate Edition, costing £74.99, offer a five-day early access period and exclusive in-game items. The game will also feature a Season Pass, granting access to four DLC releases over time, with the first available at launch. The Deluxe Edition includes the Sardaukar armor, and the Ultimate Edition encompasses all Deluxe Edition content, alongside Paul Atreides' iconic stillsuit from the 2021 film, a digital artbook, the digital soundtrack, unique colour patterns, and the Caladan Palace Set, which includes building components, placeables, and decorations.

To address player inquiries regarding the game's monetisation and post-launch support, Creative Director Joel Bylos stated that Dune: Awakening will launch as a full release and will not feature a subscription model. The game will be supported with ongoing free content updates and adhere to a traditional business model with optional DLCs. Funcom, known for its long-term support of live games like Anarchy Online and Conan Exiles, has pledged to deliver consistent updates, including new content, features, and refinements.

Dune: Awakening will launch with support for several technologies, including Nvidia's DLSS 4 with Multi Frame Generation. Compatibility for AMD's FSR and Intel's XeSS 2 technologies will also be included. Furthermore, the game will support Razer Chroma RGB, Razer Sensa HD Haptics, and Freyja Gaming Cushion.

The detailed system requirements can be found below:

Minimum

• OS: Windows 10 64-bit (or newer)
• Processor: Intel Core i5-7400, AMD Ryzen 3 1200
• Memory: 16 GB RAM
• Graphics: NVIDIA GeForce GTX 1060 (6GB), AMD Radeon 5600XT (6GB)
• Storage: 60 GB available space

Recommended

• OS: Windows 10 64-bit (or newer)
• Processor: Intel Core i7-10700K, AMD Ryzen 5 2600X
• Memory: 16 GB RAM
• Graphics: NVIDIA GeForce RTX 3070 (8GB), AMD Radeon 6700XT (12GB)
• Storage: 75 GB available space

Discuss on our Facebook page, HERE.

KitGuru says: Are you interested in Dune Awakening? Which version of the game are you planning to get?

The post Dune: Awakening PC system requirements announced first appeared on KitGuru.

Months after its initial announcement, Arm has expanded the accessibility of its Accuracy Super Resolution (ASR) upscaling technology, providing developers with more options to integrate it into games. At GDC 2025, Arm showcased pre-built and readily deployable plugins for Unity-powered games, set for release later this year. An Unreal Engine plugin is also now available. 

Built upon AMD's FidelityFX Super Resolution 2 technology, Arm claims ASR delivers a 53% increase in FPS at 2x upscaling on the Immortalis-G720 GPU (running at 2800×1260). In the “Mori” demo, we can also see ASR's capabilities using Unreal Engine 5, showcasing a 30% FPS boost while maintaining visual quality.

Arm has released a new installment in its Mali Manga series to promote ASR. The upcoming release of pre-built Unity plugins and the currently available Unreal Engine 5.3, 5.4, and 5.5 plugins simplify ASR integration for developers using these engines. This is expected to drive adoption in many popular titles, including Genshin Impact (Unity) and Fortnite (Unreal Engine).

ASR's open-source nature also allows for integration into other game engines, albeit requiring additional development effort. Moreover, it is designed to be compatible across a broad spectrum of Arm devices, regardless of manufacturer.

Discuss on our Facebook page, HERE.

KitGuru says: This new super-resolution technology fills a significant void in the Android gaming ecosystem, offering an open-source upscaling solution previously absent.

The post ARM Accuracy Super Resolution now available for Unreal Engine, Unity plugin planned too first appeared on KitGuru.

The launch of Assassin's Creed Shadows seems to have gone well. Within three days of its release and two days after reaching 1 million players, Ubisoft announced that Shadows exceeded two million. As a result, Shadows has now outperformed the launch figures of Assassin's Creed Origins and Odyssey.

On Steam, Assassin's Creed Shadows hit a new peak concurrent player count this past Sunday. According to SteamDB, Shadows' all-time peak concurrent player count on the platform was just shy of 65K, barely beating Odyssey's peak of 62K, the highest CCU on Steam for an Assassin's Creed game. Considering the 1.4M peak CCU of Monster Hunter Wilds on Valve's platform, this number isn't particularly impressive. Still, in the UK, based on GfK data (via Christopher Dring), AC Shadows is the biggest UK game launch this year, outselling Capcom's game.

The success of Shadows is a welcome boost for Ubisoft following a challenging period in 2024. The game's strong performance offers a respite after recent disappointments, such as Star Wars Outlaws and Skull and Bones. The game has received largely positive reviews, with 81.7% of Steam reviews being “positive” at the time of writing.

Amidst this success, reports indicate that Ubisoft is in discussions with Tencent regarding a potential investment in a new IP entity encompassing the Assassin's Creed series. Given their significant stake in Ubisoft, this investment would further solidify Tencent's influence over future Assassin's Creed titles. Shadows' strong launch may temporarily alleviate concerns about a full-scale ownership change at Ubisoft.

KitGuru says: Have you already tried Assassin's Creed Shadows?

The post Assassin’s Creed Shadows has already been played by 2 million users first appeared on KitGuru.

AMD is poised to introduce its Radeon RX 9060 XT graphics card in the coming weeks. Like Nvidia did with the RTX 4060 Ti and is expected to do with the RTX 5060 Ti, the new Radeon GPUs will be available with two distinct memory configurations.

While specific RX 9060 XT specifications remain undisclosed, it is suggested that the card will be available with 16 GB and 8 GB of GDDR6 memory operating on a 128-bit memory bus. Acer's listings on EEC corroborate this, but now there's another AIB partner that further solidifies this claim.

Asus has revealed its planned lineup for the RX 9060 XT (via @harukaze5719), including models within its Prime, Dual, and TUF series. These listings were discovered through the South Korean Radio Agency (1, 2), suggesting a release might be approaching. The Asus Radeon RX 9060 XT variants listed include the Prime-RX9060XT-O8G, Prime-RX9060XT-8G, Dual-RX9060XT-8G, Dual-RX9060XT-O8G, and TUF-RX9060XT-O16G-Gaming.

The absence of Radeon RX 9060 non-XT models from these listings suggests that AMD might be prioritising the launch of their higher-tier 9060 series models before introducing the cut-down versions. The new Radeon RX 9060 XT cards are expected to launch in Q2 2025.

KitGuru says: Are you awaiting for AMD's new mid-range graphics cards? Hopefully, you won't have to wait much longer.

The post Asus Radeon RX 9060 XT listings include 8GB and 16GB models first appeared on KitGuru.

More detailed specifications for Nvidia's upcoming RTX 5060 Ti graphics card, including clock speeds and a new release date, have been leaked. Moreover, some newly listed systems using the new graphics cards with 8GB and 16GB of VRAM have also been found, further suggesting that the green team's upcoming mid-range GPU will be available with two memory configurations.

According to earlier reports, the RTX 5060 Ti cards will use a GB206-300 GPU with 4,608 CUDA cores, 180W TDP, 28 Gbps GDDR7 memory, and a 128-bit memory bus. These specs will be consistent across both the 16GB and 8GB memory variants. Recently, via VideoCardz, we've learned more about the clock speeds. This report claims the base frequency is 2,407MHz, and the boost clock is 2,572MHz. This represents a 97MHz increase in base frequency and a 37 MHz increase in boost clock compared to the RTX 4060 Ti.

None of these specs have been confirmed by Nvidia yet, but the existence of two different memory configurations is becoming increasingly certain. On HP's website (via @momomo_us), the company listed two new HP Omen 16L systems using the 8 GB variant of the RTX 5060 Ti graphics card. However, at BestBuy (also via @momomo_us), there's an Omen 16L system using an RTX 5060 Ti, which is listed as having 16 GB of GDDR7 memory.

According to hardware leaker @wnxod, Nvidia has scheduled the launch of the RTX 5060 Ti variants for April 16th. This has been corroborated and expanded by HKEPC (via VideoCardz), which stated that the whole RTX 5060 series will be announced on April 15th and the release, alongside reviews, the day after. This launch date applies to both the 16GB and 8GB models.

KitGuru says: If you buy an Nvidia RTX 5060 Ti, which variant will you get? How do you think the RTX 5060 Ti will compare against its predecessor?

The post Nvidia RTX 5060 Ti now anticipated for April 16th launch first appeared on KitGuru.

Raijintek has introduced its Cyclops Lite series of AIO liquid CPU coolers, offering a blend of aesthetic appeal and efficient cooling performance. Available in black and white, the series features 240mm and 360mm radiators, catering to various system configurations.

A distinctive feature of the Cyclops Lite coolers (240mm and 360mm) is the pump block, which is adorned with a propeller-shaped design illuminated by ARGB LEDs to add some flair. The pump block incorporates a copper cold plate and a ceramic axis for the pump bearing, ensuring longevity with a rated lifespan of 50,000 hours. The radiator channels are constructed from aluminium to provide adequate heat dissipation.

Depending on the specific model, the coolers come with two or three 120mm Ageras Xtreme fans, designed for efficient airflow and customisable lighting. These fans can be daisy-chained, minimising cable clutter and simplifying installation. The Ageras Xtreme 120mm PWM fans (4-pin) use hydraulic bearings, which are rated for 40,000 hours of operation. Capable of operating within a speed range of 800 to 2,200 RPM, delivering up to 67.38 CFM of airflow at 2.57 mm H₂O static pressure while maintaining a maximum noise output of 33 dBA.

The Cyclops Lite coolers support a wide range of CPU socket types, including AM5, AM4, LGA1851, LGA1700, and LGA1200/LGA115x, ensuring compatibility with most modern platforms. Raijintek has yet to disclose pricing information for the Cyclops Lite series.

KitGuru says: Competition is fierce in the entry-level AIO segment, with brands such as Arctic and Thermalright offering good products for their prices. Do you think the Cyclops Lite can stand out from the competition?

The post Raijintek introduces new Cyclops Lite AIO liquid coolers first appeared on KitGuru.

Cherry has made a significant foray into the wireless audio segment with the launch of the Cherry Xtrfy H3 Wireless, its first multi-platform wireless gaming headset. This headset aims to bridge the gap between professional gaming audio and everyday versatility, offering a blend of gaming-optimised sound, customisable EQ settings, and multiple wireless connectivity options.

Expanding the Cherry  Xtrfy product line, the H3 Wireless builds upon its predecessors, the H1 and H2, delivering an audio experience tuned in collaboration with seasoned professional gamers. The headset's 53mm drivers are designed to amplify in-game audio cues, promising to provide an advantage in competitive scenarios.

Beyond gaming, the H3 Wireless offers dual wireless connectivity options. Users can seamlessly transition between a low-latency USB dongle, ideal for lag-free gaming, and Bluetooth, enabling effortless pairing with mobile devices. Furthermore, customisable EQ modes allow users to adjust the audio profile to suit their preferences.

Comfort was also considered when designing the H3 Wireless. The lightweight aluminium construction and oversized over-ear cups featuring soft memory foam ensure a comfortable fit for extended gaming sessions or everyday use. The integrated rechargeable battery offers up to 100 hours of uninterrupted playback, and the headset can be used while charging via USB. The Cherry Xtrfy H3 Wireless is now available for $99/€129.

KitGuru says: Cherry Xtrfy's first multi-platform wireless gaming headset shows some promise with a solid feature set, clean design, and a relatively decent price. Would you consider getting it?

The post Cherry Xtrfy launches its first multiplatform headset – the H3 Wireless first appeared on KitGuru.

Microsoft unveiled significant advancements in DirectX Raytracing (DXR) 1.2 at GDC 2025, addressing core performance bottlenecks that have traditionally hindered ray tracing adoption. Moreover, it announced that DirectX 12 will soon support neural rendering and cooperative vectors, but it should still take some time before we see them in action.

As Wccftech brought to our attention, Opacity Micromaps (OMM) and Shader Execution Reordering (SER) are the two key technologies that promise to improve ray tracing efficiency the most. OMM tackles the computational overhead associated with alpha-tested geometry by storing pre-computed opacity data, eliminating unnecessary ray-geometry intersection tests. On the other hand, SER mitigates the GPU inefficiencies caused by incoherent ray behaviour by dynamically grouping shader invocations with similar execution paths, reducing thread divergence.

These advancements represent a shift from brute-force computational approaches toward more sophisticated resource management, with Microsoft claiming performance improvements of up to 2.3x for OMM and 2x for SER. However, note that both OMM and SER are hardware-dependent features. While Nvidia has pledged support for RTX GPUs, the implementation timelines from other hardware manufacturers remain uncertain.

Furthermore, Microsoft's Shader Model 6.9 introduces cooperative vectors, a hardware acceleration design that significantly enhances matrix computation performance. This feature promises to deliver a remarkable 10x speedup in neural texture compression while reducing memory footprint by up to 75% compared to conventional methods. This can bring computationally intensive graphics within reach of mid-range consumer hardware.

The new Agility SDK's preview release, which includes the Raytracing 1.2 update, will only be available in April 2025. Considering developers will likely require months to integrate it, final products using these technologies are likely still some months away from being available to the general public.

Discuss on our Facebook page, HERE.

KitGuru says: Do you think the Raytracing update 1.2 will deliver on its promises and make the feature more accessible and less resource-intensive to systems?

The post Microsoft introduces new Raytracing 1.2 update promising up to 2.3x performance first appeared on KitGuru.

A subtle but potentially significant shift is occurring in the European graphics card market, with Nvidia quietly adjusting pricing for three of its GeForce RTX 50 series GPUs. Over at Nvidia's European stores, the company has implemented price reductions for the RTX 5090, RTX 5080, and RTX 5070 models, a move attributed to the strengthening Euro.

As VideoCardz pointed out, the US dollar's recent decline, approximately 3.9%, has likely prompted Nvidia to implement these “generous” discounts, which arrived roughly two weeks after the USD-EUR exchange rate stabilized. The current pricing situation, reflecting the change in EU MSRPs, is as follows: RTX 5090: €2329 down to €2229 (-4.3%), RTX 5080: €1169 down to €1119 (-4.3%), RTX 5070 Ti: remains at €879 (0%), and RTX 5070: €649 down to €619 (-4.6%).

The RTX 5070 Ti was the only one to maintain the MSRP, with Nvidia setting it at €879 (VAT included). This upper-mid-range offering lacks a Founders Edition, leaving board partners responsible for producing MSRP-compliant alternatives. Reports suggest that some retailers struggle to move RTX 50 series inventory due to inflated prices, with consumers hesitant to purchase.

While Nvidia's price adjustments, ranging from 4.3% to 4.6%, are a welcome development, their impact across the Eurozone remains to be seen. Given the relatively modest reductions, it's uncertain whether they will significantly stimulate demand.

Discuss on our Facebook page, HERE.

KitGuru says: Do you think prices will decrease in European stores now that the MSRPs have been reduced? Even if they drop, would you become more inclined to get one of these RTX 50 series GPUs?

The post Nvidia RTX 50 graphics cards are now cheaper across Europe first appeared on KitGuru.

The Nvidia RTX 5060 series launch has reportedly been pushed back, casting uncertainty over the release timeline for the lower-tier Blackwell gaming GPUs. Prior reports claimed that a launch date in April had been communicated to board partners. However, new information suggests a delay has taken place. 

Information taken from the Board Channels forum (via HXL) suggests that Nvidia has recently informed the board partners of a revised delivery schedule for the RTX 5060 series. The phrasing implies a delay, although Nvidia's lack of official confirmation prevents a formal declaration of such.

According to this information, the RTX 5060 Ti, expected to become available in both 16 GB and 8 GB formats, is now planned to debut in mid-April, while the RTX 5060 is slated for a mid-May release. Note that these dates are subject to change, as indicated by the forum's post wording.

The rumoured specifications for these cards remain unaltered. As such, expect the RTX 5060 Ti to feature 4,608 cores across both memory variants, with a 180W TDP. As for the RTX 5060, it's projected to have 3,840 cores and an 8 GB memory configuration. Additionally, reports indicate that Nvidia is developing an RTX 5050, targeting the entry-level segment.

Discuss on our Facebook page, HERE.

KitGuru says: Are you waiting for the release of the RTX 5060 series to upgrade your GPU? Which model are you interested in?

The post Rumoured RTX 5060 launch delayed first appeared on KitGuru.

The relentless pursuit of memory overclocking records has witnessed another milestone. Renowned overclocker HiCookie achieved a new DDR5 world record, pushing the boundaries of memory speed to an astounding 12,752 MT/s (DDR5-12752).

This feat was achieved using the Gigabyte Z890 Aorus Tachyon Ice motherboard, a 24 GB V-Color Manta XFinity RGB DDR5 memory stick (68-127-127-127-2T timings) cooled with liquid nitrogen, and an Intel Core Ultra 9 285K CPU running with only two cores (without HT) at 420 MHz. The validations at CPU-Z and HWBOT cement HiCookie's position at the pinnacle of memory frequency overclocking, at least for now.

The memory speed record has been a fiercely contested arena in recent months, with numerous overclockers fighting for the top spot. The proximity of Splave's second-place achievement, reaching DDR5-12735, shows how competitive this pursuit has become, so we wouldn't be surprised if someone breaks the newly set record soon.

Discuss on our Facebook page, HERE.

KitGuru says: Seeing how often the DDR5 memory speed world record has been shattered in the last couple of years, how long will this one last? Do you think we'll soon see someone breaking the DDR5-13000 barrier?

The post Overclocker breaks DDR5 world record with 12572MT/s first appeared on KitGuru.

Microsoft has launched the Xbox Adaptive Joystick, a new peripheral designed to enhance accessibility for gamers with varying mobility needs. This wired controller, exclusively available through the Microsoft store, features seven fully remappable buttons and connects directly to Xbox consoles, PCs, or the Xbox Adaptive Controller.

This joystick represents the latest addition to Microsoft's Adaptive lineup, a series of controllers specifically engineered to address the requirements of gamers with limited mobility. The device's thumbstick can be customised with 3D-printed attachments, offering a level of personalisation previously unavailable in mainstream gaming peripherals. With limited options from other major manufacturers, the Xbox Adaptive Joystick has the potential to become an essential device for an often-overlooked segment of the gaming community.

While the Xbox Adaptive Joystick was initially announced last year, its official release has only now arrived, bringing the much-anticipated 3D-printed accessory support. Currently, customisation is focused on thumbstick toppers, which can be personalised through the Xbox Design Lab studio. Six base shapes are available, which users can customise in height and width to fit their needs. The 3D model file is free.

The seven remappable buttons, customisable through Microsoft's software, allow for a single stored preset on the device. The button layout includes four front-facing buttons, bumper and trigger options on the rear, and a clickable thumbstick button. The device can be used with the Xbox Adaptive Controller or independently, though full-button functionality requires a separate Xbox controller. The Xbox Adaptive Joystick is now available for £24.99/$29.99, exclusively through Microsoft.

KitGuru says: The new Xbox adaptive joystick is relatively cheap compared to other adaptive controllers. However, it's more limited than most, making it look more like an accessory than a standalone product.

The post Microsoft launches new nunchuk-like Xbox adaptive joystick first appeared on KitGuru.

At GTC 2025, Nvidia has shared its roadmap for data centre GPUs, revealing a series of advancements promising to reshape the AI and HPC landscapes. The company's vision extends beyond the immediate future, encompassing refinements to its Blackwell architecture, introducing the Rubin (and Rubin Ultra) platform and unveiling a brand-new architecture codenamed Feynman.

The Blackwell Ultra NVL72, slated for release in the latter half of this year, marks the first step in this evolution. This system will use 72x Blackwell Ultra chips, each featuring GPUs and 288 GB of HBM3e memory, delivering a claimed 1.5x performance boost over the existing GB200 NVL72 platform. The complete rack, which also includes 36 Arm Neoverse-based Nvidia Grace CPUs, will boast 20 TB of HBM memory and 40 TB of fast memory.

Images via Nvidia

Looking further ahead, Nvidia has detailed its Rubin and Rubin Ultra platforms. Rubin, the first architecture built on Nvidia's next-generation design principles, is expected to debut in the second half of 2026. The company is developing the Rubin NVL144 platform, which will house Rubin chips featuring two GPUs and 288 GB of HBM4 memory. This platform will also pack 75 TB of fast memory Vera CPUs, each with 88x Custom ARM cores with dual threads. This whole suite promises to deliver up to 3.3x the performance of the Blackwell Ultra NVL72 platform.

Built on top of the Rubin NVL144, Nvidia plans to launch the Rubin Ultra NVL 576 with 365 TB of fast memory in H2 2027. This platform will pack the same Vera CPUs as the Vera Rubin NVL144, but instead of the standard Rubin chips, it will use Rubin Ultra, featuring four GPUs and 1 TB of HBM4e memory. As such, one can expect up to 14x more performance than the Blackwell Ultra NVL72 platform.

Lastly, Nvidia has provided a glimpse into its long-term vision with the Feynman architecture, which is anticipated to arrive in 2028. Details are still scarce, but we learnt it will leverage next-generation HBM memory technology.

KitGuru says: Do you think Nvidia will use any of these new architectures for products other than those aimed at HPC and the data-centre segments?

The post Nvidia unveils new data centre GPU roadmap at GTC 2025 first appeared on KitGuru.

Razer has unveiled Wyvrn, a comprehensive developer platform integrating AI to streamline game development and enhance player experiences. This platform introduces a suite of tools, including the AI QA Copilot for automated quality assurance and the AI Gamer Copilot for real-time in-game assistance, alongside updates to Razer's Chroma RGB, THX Spatial Audio, and Sensa HD Haptics technologies.

The AI QA Copilot, a cloud-based plugin compatible with Unreal Engine, Unity, and custom engines based on C++, promises to revolutionise quality assurance testing. By leveraging AI, this tool aims to reduce the time and cost associated with manual testing. The AI assistant monitors gameplay, automatically detects errors and performance issues, and generates detailed QA playtest reports, incorporating AI-detected and human-identified flaws. Razer claims the AI can identify 20-25% more defects than manual testing, potentially reducing QA time by up to 50% and costs by up to 40%. The AI QA Copilot is designed to learn from identified defects, continuously improving its accuracy and efficiency.

Another key component of Wyvrn is the AI Gamer Copilot, an AI-powered voice assistant that provides real-time in-game guidance. This tool offers tactical advice for competitive multiplayer games and strategies for overcoming challenges in single-player titles. Removing the “Project Ava” naming, as announced at CES 2025, suggests the AI Gamer Copilot is nearing its official release, but a specific date hasn't been shared.

Beyond AI-driven tools, Wyvrn also enhances Razer's existing technologies. Razer Sensa HD Haptics now features sim racing game integration through collaboration with SimHub, enabling immersive haptic feedback in titles like Assetto Corsa Evo for users of the Freyja haptic chair cushion and Kraken V4 Pro headset. Razer reports Sensa haptics are now supported in over 100 titles, with further expansion planned.

Razer Chroma RGB is also getting some love, introducing new 3D reactive lighting effects. Additionally, the company is releasing a THX Spatial Audio Plus plugin for Wwise, an open-source format designed to deliver more realistic 3D audio in games compared to proprietary solutions.

KitGuru says: What kind of experiences do you expect developers will create using Razer Wyvrn?

The post Razer gets into AI with new Wyvrn ecosystem first appeared on KitGuru.