RSS Reader

OpenCV 4.13 Brings More AVX-512 Usage, CUDA 13 Support, Many Other New Features

Phoronix | Dec. 31, 2025, 6:35 a.m. | Original

OpenCV 4.13 is out this New Year's Eve in providing the latest open-source computer vision (CV) capabilities. OpenCV 4.13 brings a wide variety of enhancements to this widely-used computer vision library.

With OpenCV 4.13 there are several performance optimizations for leveraging this computer vision code under Windows on ARM. There are also some test failure fixes for Windows on ARM. Meanwhile on the x86_64 side, there is continuing to be more usage around

Shotcut 25.12 Released With 10-bit Video CPU Pipeline, Linear Color Processing

Phoronix | Dec. 30, 2025, 8:21 p.m. | Original

December happens to be a busy month for video editor releases in the open-source world. This month there's been the release of Flowblade 2.24 , OpenShot 3.4 , Kdenlive 25.12 , and now there is Shotcut 25.12 before closing out the month and year.

Shotcut 25.12 is now available for this cross-platform free software video editor bjuilt atop the MLT Multimedia Framework and FFmpeg. With Shotcut 25.12 there is now

Unexpected Surprise: Windows 11 Outperforming Linux On An Intel Arrow Lake H Laptop

Phoronix | Dec. 30, 2025, 5:04 p.m. | Original

Typically when receiving any review hardware preloaded with Microsoft Windows I tend to run some Windows vs. Linux benchmarks just as a sanity test plus it still seems to generate a fair amount of interest even though the outcome is almost always the same: Linux having a hefty performance advantage over Windows especially in the more demanding creator-type workloads. As an unexpected twist and time consuming puzzle the past two months, when recently testing out the Lenovo ThinkPad P1 Gen 8 it's faster for numerous workloads now on Microsoft Windows 1

X.Org IMAKE Updated For Those Not Yet Transitioned To Autoconf/Automake Or Meson

Phoronix | Dec. 30, 2025, 3:32 p.m. | Original

X.Org package wrangler Alan Coopersmith at Oracle announced today the release of imake 1.0.11, the newest version of this utility that 20+ years ago was used extensively as part of the X Window System build process for generating Makefiles from a template. With this first imake point release in two years, imake itself can now be built via Meson and there is now support for RISC-V and LoongArch architectures.

Over the past twenty years the still-active X11 / X Window System

Some Meaningful Performance Benefits For Clang + LTO Built Linux Kernels

Phoronix | Dec. 30, 2025, 1:50 p.m. | Original

Over the past few years building the Linux kernel with Clang has matured a lot thanks to upstream improvements to both LLVM/Clang and the Linux kernel. As it's been a while since our last comparison for GCC vs. Clang built kernels on the resulting system performance, our latest year-end 2025 benchmarking is providing a fresh look at the Linux 6.19 upstream Git kernel built under the latest stable GCC 15 and LLVM Clang 21 compilers. Plus with the Clang-built kernel

Intel's Xe Linux Driver Ready With Multi-Device SVM To End Out 2025

Phoronix | Dec. 30, 2025, 10:11 a.m. | Original

Intel's open-source graphics driver engineers are ending out 2025 with a bang. Sent out today was the final drm-xe-next pull request of the year of new feature material ready for the next version of the Linux kernel. Today's pull adds support for SR-IOV scheduler groups as well as multi-device Shared Virtual Memory (SVM) support.

The DRM-Xe-Next pull request is on its way to DRM-Next as code to queue in advance of the

LLVM 22 Lands NVIDIA Olympus CPU Scheduling Model

Phoronix | Dec. 30, 2025, 9:46 a.m. | Original

NVIDIA's Olympus are the ARM64 cores found within the upcoming Vera CPU that will be paired with Rubin. Olympus cores are claimed to be twice as fast as NVIDIA's current CPU cores found in Grace and based on Neoverse-V2. Earlier this year the open-source compilers landed initial support for Olympus while now a proper CPU scheduling model has been upstreamed into LLVM 22.

The initial NVIDIA Olympis compiler enablement that landed earlier this year for the GCC and LLVM/Clang compilers confirmed that it

Linux 6.19 Kernel Benchmarks With X86_NATIVE_CPU Optimization

Phoronix | Dec. 30, 2025, 7:27 a.m. | Original

Added to the Linux kernel earlier this year was the new X86_NATIVE_CPU Kconfig option to enable compiler optimizations for the local/native CPU in use when building the Linux kernel. In effect about ensuring that the "-march=native" compiler flag is set for the kernel build for optimizing the Linux kernel build for your processor being used. Back with Linux 6.16 I ran some benchmarks of the Linux kernel build with X86_NATIVE_CPU to gauge the impact. Now

InputPlumber 0.70 Released With Expanded Hardware Support

Phoronix | Dec. 30, 2025, 6:55 a.m. | Original

InputPlumber 0.70 is out today as the newest feature update to this open-source input router and re-mapper daemon for Linux systems. With more gaming handhelds coming to market and other controllers as well as the upward trajectory of Linux gaming, InputPlumber is becoming more applicable for this daemon to combine various input devices into different virtual device formats.

Today's InputPlumber 0.70 release expands hardware support with now handling the OneXPlayer X1 series and improving the Anbernic Win600 support. The OneXPlayer

XWayland Gets Patched For Incorrect Pointer Coordinates

Phoronix | Dec. 30, 2025, 6:40 a.m. | Original

An important fix has made it into the X.Org Server XWayland codebase ahead of the new year. XWayland has been fixed to avoid sending incorrect pointer coordinates to X11 clients on pointer enter events.

A fix was merged recently to the XWayland code to avoid sending incorrect pointer coordinates to X11 clients on pointer enter events, namely around the cursor/input returning to an XWayland window. The patch now ensures that the pointer sprite coordinates are correctly updated when the pointer enters an XWayland window.

The fix came

Menu