GoKawiil
Tech News
YouTuber Valery Kryzhanovsky used the MAME emulator with ReShade and RenoDX to force Nvidia's leaked DLSS 5 upscaler onto the MS-DOS version of Pac-Man, applying a 4x Multipass technique across four rendering modes: Natural, Default, Cinematic, and Denoise. The results were glitchy, with some modes erasing Inky's eyes and the game's dots, since Pac-Man lacks the 3D depth data DLSS 5 is built to process.
Valery Kryzhanovsky has demonstrated the use of Nvidia's DLSS 5 technology on vintage games like the 1991 DOS shooter Duke Nukem and the 1972 game Pong. His experiments involve running these titles through DOSBox and overlaying DLSS 5 via ReShade, resulting in enhanced visuals with added depth and object detection. He provides detailed instructions and downloadable tools for others interested in replicating the process.
A new open-source project called OpenDLSS reimplements the neural network behind Nvidia's DLSS 5, reproducing the same 71-block transformer architecture used in DLSS-NR build 310.8.0, down to matching intermediate outputs at all 75 block boundaries. The project runs on Vulkan using FP8 tensor-core computation, and includes a separate WebGPU browser port that reproduces identical results without tensor cores or FP8 support. Users must supply their own copy of Nvidia's trained model weights to run it.
An open-source GitHub project called DLSS-NR-on-AMD is reimplementing Nvidia's DLSS 5 Neural Rendering runtime to work on AMD Radeon RX 9000 and RX 7000 series cards, currently supporting DirectX 12 titles that use FSR, including Cyberpunk 2077 and GTA V Enhanced. Early builds saw frame rates collapse from over 80 FPS to around 11-12 FPS on a Radeon RX 9070 XT, but successive alpha releases have delivered roughly 74% combined performance gains within 24 hours, and a new AMDNR launcher now automates installation.
Korean outlet QuasarZone tested the GeForce RTX 5090 Founders Edition running Cyberpunk 2077 at 4K with DLSS 5 enabled, recording peak power draw near 650W and a 16-pin (12VHPWR) connector temperature of 91.7 degrees Celsius, higher than the GPU die itself. The RTX 5080 also showed a temperature rise with DLSS 5 on, though smaller in magnitude than the 5090's.
Microsoft has expanded Automatic Super Resolution, its NPU-driven upscaling feature built into Windows, to Copilot+ PCs powered by Intel's Core Ultra Series 3 chips, a capability previously exclusive to Qualcomm Snapdragon X devices. The feature aims to boost frame rates in games rendered at up to 1080p, with Microsoft citing gains of up to 15% in titles like Assassin's Creed Valhalla, targeting a 60 FPS threshold.
CyberPowerPC's SLC8500WST prebuilt pairs AMD's Ryzen 7 9800X3D processor with Nvidia's RTX 5080 Founders Edition graphics card, 32GB of DDR5-6000 memory, and a 1TB PCIe 4.0 SSD, all for $2,299 at Walmart. The CPU and GPU alone typically cost over $2,000 when bought separately, making the bundled memory and storage essentially free by comparison.
A GitHub user known as Uzbekunknown built a project called "dlss-nr-on-intel" that reimplements Nvidia's DLSS 5 71-block U-Net neural network to run on Intel's Xe2 architecture via the Vulkan cooperative matrix extension, rather than wrapping the original DLL. The code was largely written by AI assistants Claude and GPT-6 Astra, with the developer directing the process, and testing on Intel's Arc 140V graphics showed the technique working but extremely slow, around 10.5 fps at 360p in Tekken 7.
A developer known as MAAN built a browser-based demo that reimplements Nvidia's DLSS 5 neural rendering network using WebGPU compute shaders, hosted on Cloudflare Workers with adjustable settings and sample scenes. The port reportedly works on non-Nvidia GPUs and even macOS, using roughly 147MB of extracted model weights alongside a small JavaScript runtime, though each rendered frame currently takes one to two seconds. MAAN says the weights came from a leaked DLSS 5 library file and plans to release the source code on GitHub soon.
Leaker Kepler_L2 claims AMD's upcoming RDNA 5 graphics architecture will include its own neural lighting technology, similar to Nvidia's DLSS 5. Because next-gen consoles are expected to use AMD chips, this feature could bring neural rendering to the PS6 and the next Xbox rather than staying limited to high-end PC cards.
NVIDIA showcased DLSS 5 at GTC using footage from Resident Evil Requiem and Starfield, but viewers noticed the upscaling gave characters smoothed, glossy skin resembling a beauty filter rather than photorealism. Artists including Karla Ortiz criticized the technology for overriding developers' intentional art direction, while Capcom reportedly didn't know its game was used in the demo. CEO Jensen Huang dismissed the criticism as 'completely wrong' during a GTC Q&A, further fueling online anger.
A modder known as Perseval_BLR, aided by Claude AI, built an app called NeuralScreen that overlays Nvidia's leaked DLSS 5 neural rendering technology onto any windowed application, not just games. Demonstrations show it applied to YouTube gameplay footage, a scene from Interstellar, and Doom 3 running in a window, with customizable presets and key-bound controls for switching, recording, and comparisons.