Intel has shifted the blame for a lack of Arc GPU in Crimson Desert back on developer Pearl Abyss, saying it's reached out to the studio "many times" over the past several years.

Fortnite's new season not only brings new content for the players but also gives them a chance to earn real-life rewards, such as an RTX 5080. If you're good enough at the game and win enough "rivalries" in matches every week and rank among the top five players in the world, Epic will give you a free RTX 5080.

AMD has added Ray Regeneration 1.1 and FSR 4.1 upscaling to its RDNA 4 GPUs, bringing it to parity with Sony's PSSR 2 on the PS5 Pro. Games that support these features will have better ray tracing quality with more accurate shadow detail, while also getting a sharper-looking image through ML-based upscaling.

Despite appearing intact, the GPU was missing critical components internally, making the scam difficult to detect without disassembly.

Walmart stocks up on Nvidia's GeForce RTX 40-series (codenamed Ada Lovelace) graphics cards amid an AI-driven memory shortage.

Optiscaler has provided a new update that helps remove ghosting on the INT8 version of FSR 4 for RX 6000 series graphics cards.

Following in Nvidia's footsteps, Intel has now officially adopted Microsoft's Advanced Shader Delivery to make shader compilation much faster in games. Intel is calling it Precompiled Shader Distribution and it's available on a bunch of Arc GPUs right away, supported in 11 games at launch, with more likely to follow. AMD is now the only company left who hasn't officially embraced this feature.

Nvidia publishes 2026 – 2028 data center roadmap with Rosa CPU, Feynman GPU, optical NVLinks and Groq LPUs with NVFP4 and NVLink.

Nvidia CEO Jensen Huang responded to backlash against DLSS 5, saying artistic control remained with developers and that the AI works with existing geometry.

Nvidia's DLSS 5 AI model uses a deep awareness of environmental lighting and how that light interacts with various materials in a scene to dramatically upgrade the appearance of games, and the results can be both impressive and uncanny.

Nvidia shows off its next-generation Kyber rack-scale solution to be powered by Rubin Ultra GPUs with four compute chiplets and 1 TB of HBM4E memory per package.

Nvidia's slides at GTC lack any mentions of Rubin CPX, but praise Groq LPUs instead.

Nvidia believes it earn $1 trillion by selling AI hardware, says co-founder and chief executive Jensen Huang.

Six commercial space companies are understood to have already deployed the platform.

At GTC 2026, Nvidia revealed the Groq 3 accelerator and Groq LPX rack as part of the Vera Rubin platform. These SRAM-packed, inference-focused chips deliver large amounts of memory bandwidth to help Rubin deliver low-latency interactions with AI models spanning trillions of parameters and million-token contexts.

Nvidia announced more details about its new 88-core Vera data center CPUs, claiming impressive 50% performance gains over standard CPUs, fueled by a 1.5X increase in IPC from its Olympus cores. The firm also unveiled its new Vera CPU Rack architecture, which brings 256 liquid-cooled CPUs into one rack for CPU-centric workloads.

A single-fan RTX 4070 Ti Super had been in the works at Zephyr, a Chinese vendor, for a while, and it was close to completion, with even thermal testing data publicly released. Unfortunately, the memory crisis has gotten to Zephyr as well, and it has cancelled the project, choosing to instead develop an RTX 4070 Super instead.

An RTX 5070 Ti with user-applied liquid metal died because the TIM leaked out everywhere and shorted multiple components, eventually killing the core as well. Despite being part of a "repair" video, there's nothing really here to fix, as most of the important ICs would need to be replaced or at least reballed.

A new LLVM patch has added V_FMA_F32, a 3-operand fused multiply-add (FMA instruction and introduced the VOPD3 instruction format for RDNA 5. Both of these changes should make it easier for compilers to use dual issue execution, working around the strict pairing rules that would otherwise limit max FP32 throughput in certain workloads.

Meanwhile, the older 7G105 GPU name no longer appears on the site; all four cards now sit under the LX product branding.

At GDC 2026, Nvidia held a presentation somehow aimed at gamers and not data center clients. Still, it was sprinkled with AI pat-on-the-backs, with the company touting that its future gaming GPUs will offer 1,000,000 better path tracing performance. And the current-gen Blackwell family is apparently already 100,000x better due to dedicated Tensor and RT cores.

Nvidia says it is permissible for ByteDance to use AI clusters outside of China to develop its AI prowess as long as these clusters are built in compliance with the U.S. export controls.