Language Selection

English French German Italian Portuguese Spanish

Linux Graphics and Games

Filed under
Graphics/Benchmarks
Gaming
  • Marek Has Been Taking To AMDGPU LLVM Optimizations

    Well known AMD open-source driver developer Marek Olšák has ruthlessly been optimizing the Radeon Mesa driver stack for years. With RadeonSI getting fine-tuned and already largely outperforming the AMDGPU-PRO OpenGL driver and most of the big ticket improvements complete, it appears his latest focus is on further optimizing the AMDGPU LLVM compiler back-end.

    This AMDGPU LLVM compiler back-end is what's used by RadeonSI but is also leveraged by the RADV Vulkan driver, among other potential use-cases. Lately Marek has been filing patches for optimizing the instructions generated during the shader compilation process.

  • FFmpeg Expands Its NVDEC CUDA-Accelerated Video Decoding

    A few days back I wrote about FFmpeg picking up NVDEC-accelerated H.264 video decoding and since then more FFmpeg improvements have landed.

    As mentioned in the earlier article, NVDEC is the newer NVIDIA video decoding interface that is succeeding their Linux-specific VDPAU in favor of the cross-platform, CUDA-based NVIDIA Video Codec SDK. There's also NVENC on the video encode side, while the recent FFmpeg work has been focused on the NVDEC GPU-based video decoding.

  • Intel Batch Buffer Logger Updated For Mesa

    Intel's Kevin Rogovin has been working on a "BatchBuffer Logger" for the Intel graphics driver that offers some useful possibilities for assisting in debugging/analyzing problems or performance penalties facing game/application developers.

    The BatchBuffer Logger is designed to allow correlating API calls to data that in turn is added to a batch buffer for execution by the Intel graphics processor. The logger additionally keeps precise track of the GPU state and can report various metrics associated with each API call.

  • Feral Interactive continues to contribute to Mesa as one of its developers gets commit privileges

    Alex Smith of Feral has been granted the right to push code into Mesa, a continuing sign of the commitment of Feral to Mesa and Vulkan.

    In this recent exchange Feral dev and active Mesa contributor, Alex Smith, has asked and gotten permission to create an account to directly access the Mesa driver’s git. His stated purpose is to provide fixes for Vulkan drivers, so we can take that as a sign that Feral is pretty serious at not only contributing to the open source Mesa project but also at using the Vulkan API in their current and future ports.

  • I played a bit of Die Totenmaske and it was a very strange experience
  • Valve has added 11 new currencies to the Steam Store

    Recent changes to the Steam Store have seen the addition of more local currencies for customers in different regions. Expect to get more bang for your Peso or Dinar.

    Originally tweeted by the excellent SteamDB, it would seem that customers in different regions will be able to buy from the Steam Store using their local currency. This usually means lower prices and no fiddling about with conversion rates for currency and prices are also adjusted for regional standards. The changes went live earlier and users in the affected countries have gotten emails telling them about the new changes to the Steam Store.

  • Trüberbrook, a beautiful adventure game with Linux support, is on Kickstarter

    This story-driven adventure game set in 1960s Germany places you in the role of a young scientist who finds himself having to save the world.

  • Intra-System: Trust Issues fiendishly encourages betrayal, my thoughts on the experience

More in Tux Machines

Funding for Open 'Core' Companies

'Proper' GNU/Linux on Google OSes

  • Google’s Fuchsia OS will support Linux apps
    Google’s non-Linux-based Fuchsia OS has added an emulator for running Debian Linux apps. Like its upcoming Linux emulator for Chrome OS, Fuchsia’s “Guest” app will offer tighter integration than typical emulators. Google has added a Guest app to its emergent and currently open source Fuchsia OS to enable Linux apps to run within Fuchsia as a virtual machine (VM). The Guest app makes use of a library called Machina that permits closer integration with the OS than is available with typical emulators, according to a recent 9to5Google story.
  • Here are the latest Chrome OS devices that will support Linux apps
    The ability to run Linux apps in virtual machines in Chrome is expanding beyond Google's flagship Pixelbook line of Chromebooks. The feature, for which plans were first discovered in late February, was formally announced by Google at I/O 2018. Unlike the existing solution, Crouton, support for Linux apps does not require enabling developer mode on Chrome OS, allowing users to install Linux apps without needing to sacrifice security protections. In addition to the Pixelbook, support for the new Crostini virtual machine feature has also come to the original Samsung Chromebook Plus, the detachable HP Chromebook X2, and the ASUS Chromebook Flip C101. Likewise, according to a report from xda-developers, the feature is coming to the Acer Chromebook Spin 13 and Chromebook 13, as well as 2018-era Chromeboxes, which all share the same board ID "fizz." Of these, the Acer Chromebox CX13 series and ASUS Chromebox 3 series both have multiple SKUs, maxing out with an Intel Core i7-8550U paired with 16GB RAM and 64GB storage for $750.
  • Linux App Support Is Coming To Acer Chromebook Flip C101
    Acer’s Chromebook Flip C101 is now officially the latest Chrome OS device expected to be in-line for virtualized Linux app support, following a new commit pushed to the Chromium Gerrit on June 15. That places the Flip C101 in a very select club alongside Google’s Pixelbook, the HP Chromebook x2, and the first generation Samsung Chromebook Plus. Of course, there’s no official date with regard to when Linux App support will arrive for the Chromebook Flip C101. If previous trends are followed, then it shouldn’t take too long at all for its official arrival in the Canary Channel of the OS. That comes following a commit indicating that support for the new feature has been moved from the Samsung Chromebook Plus to the devices’ shared parent board. Since only the Chromebook Plus and Chromebook Flip C101 share that board, dubbed “Gru,” that suggests that both devices will support Linux apps in a virtual environment.

Linux Foundation: New Study, Automotive Grade Linux (AGL), and Hyperledger Fabric

Graphics: AMDGPU, Nvidia, Apple's Harm to Science

  • AMDGPU DRM Driver To Finally Expose GPU Load Via Sysfs
    The AMDGPU DRM driver appears to finally be crossing the milestone of exposing the current GPU load (as a percentage) in a manner that can be easily queried via sysfs. For years I've been frustrated via the lack of standardization of sysfs/debugfs files among the DRM drivers and some seemingly basic information not being exposed in such a manner that easily benefits various desktop plug-ins, those wanting to script basic monitors/checks/etc around such outputs, and use-cases like with the Phoronix Test Suite for easily querying this information too for its sensor recording. One of the frustrations with the Radeon Linux stack has been that there wasn't a trivial way to read the GPU load usage as a percentage... There's been ways if installing third-party utilities like RadeonTool, but no universal solution nor one that doesn't require root and would be widely available.
  • Radeon Software 18.20 Stable Released With Official Ubuntu 18.04 LTS Support
    The Radeon Software "AMDGPU-PRO" 18.20 hybrid driver stack is now available with official support for Ubuntu 18.04 LTS and Red Hat Enterprise Linux 17.20. Two months after the debut of the Ubuntu 18.04 LTS "Bionic Beaver" release, the Q2'2018 Radeon Sotware for Linux driver update has arrived with support for this latest long-term support release. Radeon Software 18.20 was officially released last week but seemingly went under everyone's radar until now.
  • Nvidia Releases a Batch of Open Source Tools for AI
    Graphics processors increasingly used as hardware accelerators for deep learning applications are also being deployed with the Kubernetes cluster orchestrator as another way to accelerate the scaling of training and inference for deep learning models. The two-front approach includes Nvidia’s (NASDAQ: NVDA) release to developers this week of a Kubernetes on GPU capability aimed at enterprises training models on multi-cloud GPU clusters. Previously, Google (NASDAQ: GOOGL) launched a beta version of GPUs on its Kubernetes Engine aimed at accelerating machine learning and image processing workloads.
  • AI caramba! Nvidia devs get a host of new kit to build smart systems
    Nvidia has released a bunch of new tools for savvy AI developers in time for the Computer Vision and Pattern Recognition conference in Salt Lake City on Tuesday.
  • Chemists criticise mooted shutdown of 3D visualisation tools
    End of support for Apple’s OpenGL programming interface could pull the plug on molecular modelling software Researchers are voicing concerns over a move that may affect many 3D visualisation programs that are commonly used in computational research. Apple’s Macintosh operating systems (macOS) is set to end support for OpenGL, the programming interface frequently used to display 3D graphics in medical and scientific visualisation software, which has existed since 1992. Nearly all open source and commercial chemistry visualisation programs that are used to display atoms, molecules, bonds and protein ribbons – such as Mercury, VMD and PyMOL – use the system.