Vue normale

Reçu — 21 avril 2026 Actualités numériques

AMD Ryzen 9 9950X3D2 Benchmarks: The Best Desktop Performance For Linux Developers, Creators

21 avril 2026 à 13:00
Today we can finally share performance benchmarks of the long-rumored AMD Ryzen 9 9950X3D2 Dual Edition processor. This new halo product for the Ryzen 9000 series desktop line-up offers captivating performance for developers frequently compiling code, creators, technical computing workloads for students or hobbyists or those not able to afford a Threadripper / EPYC type workstation, or similar heavy computing use. With the 16 cores / 32 threads and both CCDs having 3D V-Cache, the AMD Ryzen 9 9950X3D2 offers leading performance among current generation desktop processors.

RADV Driver Enables Host Image Copy By Default For RDNA2 & Newer

21 avril 2026 à 12:24
Introduced back in 2023 with Vulkan 1.3.258 was VK_EXT_host_image_copy to copy data between host memory and images on the host processor without needing to stage the data through a CPU-accessible buffer. This direct CPU-to-GPU image data transfer path can reduce memory usage during asset loads and all around more efficiency and performance. Finally now the RADV open-source Radeon driver is enabling support by default...

AMD's GAIA Makes It Easier To Import/Export Custom AI Agents Across PCs

21 avril 2026 à 00:36
AMD on the software side continues investing heavily in GAIA "Generative AI Is Awesome" as their cross-platform solution built around the Lemonade SDK for running local AI agents on your AMD-powered hardware from CPUs to GPUs and NPUs. With today's GAIA update, custom-generated AI agents are now portable with easy import and export support...

HarfBuzz Continues Improving Its New GPU-Accelerated Text Shaping Library

21 avril 2026 à 00:23
Released at the beginning of the month was a new version of HarfBuzz, a widely-used, open-source text shaping engine. With this HarfBuzz 14.0 release it introduced a GPU-based text rasterization library that supported GLSL shaders as well as HLSL, WGSL, and APple's Metal MSL. Since then this GPU-accelerated library has been seeing more improvements...
❌