Language Selection

English French German Italian Portuguese Spanish

Graphics/Benchmarks

NVIDIA 435.27.06 Vulkan Linux Driver

Filed under
Graphics/Benchmarks
Linux
  • NVIDIA 435.27.06 Vulkan Linux Driver Has Useful Display Improvements

    Released on Wednesday was the NVIDIA 435.27.06 Linux driver as their newest beta build focused on offering better Vulkan driver support.

    While the NVIDIA 435.27.06 driver doesn't offer any new Vulkan extensions or landmark new features, it does offer some practical improvements. Catching our eye in particular is the better fullscreen exclusive support handling for non-primary monitors and better G-SYNC support for borderless windows. The other fixes are also quite useful too.

  • NVIDIA 435.27.06 Vulkan Beta driver is out

    If you like to live on the wild side, NVIDIA yesterday released a brand new update to their Vulkan Beta driver series with version 435.27.06.

The Firefox + Chrome Web Browser Performance Impact From Intel's JCC Erratum Microcode Update

Filed under
Graphics/Benchmarks

With yesterday's overview and benchmarks of Intel's Jump Conditional Code Erratum one of the areas where the performance impact of the updated CPU microcode exceeding Intel's 0~4% guidance was on the web browser performance. Now with more time having passed, here are more web browser benchmarks on both Chrome and Firefox while comparing the new CPU microcode release for the JCC Erratum compared to the previous release. Simply moving to this new CPU microcode does represent a significant hit to the web browser performance.

In this article is just a look at how the updated CPU microcode for the JCC Erratum affects the Mozilla Firefox and Google Chrome web browser performance. This article isn't looking at any impact from the also new Zombieload TAA mitigation (that's coming in a separate article shortly) or anything else but simply benchmarking both of these web browsers with the old and new CPU microcode on a Skylake-X system.

Read more

Graphics: Adreno, Vulkan, Khronos, and Radeon Pro Software for Enterprise

Filed under
Graphics/Benchmarks
  • Qualcomm's Adreno 640 GPU Is Working Easily With The Freedreno OpenGL/Vulkan Drivers

    The Adreno 640 GPU that is used by Qualcomm's Snapdragon 855/855+ SoCs is now working with the open-source Freedreno Gallium3D OpenGL and "TURNIP" Vulkan drivers with the newest Mesa 20.0 development code.

    Besides the forthcoming Adreno 680/685 GPUs for Snapdragon-powered Windows laptops, the Adreno 640 is at the top of the Adreno 600 series line-up. The Adreno 640 is 7nm based and has more ALUs than the Adreno 630 and older, an 899~1037 GFLOPS rating, and other improvements.

  • Intel's Vulkan Linux Driver Lands Timeline Semaphore Support

    A change to look forward to with Mesa 20.0 due out next quarter is Vulkan timeline semaphore support (VK_KHR_timeline_semaphore) for Intel's "ANV" open-source driver.

    Vulkan timeline semaphore support is the latest synchronization model for the Vulkan graphics API and building upon earlier primitives. The Vulkan Timeline Semaphore extends VkSemaphore and supports signal/wait from host threads, better platform support, a monotonically increasing counter than can be used for more descriptive purposes, and other design improvements.

  • Khronos Next Pursuing An Analytic Rendering API

    The Khronos Group has been expanding into a lot of new areas in recent times from OpenXR to 3D Commerce to NNEF and now forming an exploratory group for creating an analytic rendering API.

    The Khronos Analytic Rendering API would be an industry standard API around data visualizations. This API would be a step above graphics APIs like Vulkan and be catered to data presentation purposes. The API has yet to be formalized as it's still in the early stages but would likely be akin to a vendor-neutral equivalent of NVIDIA VisRTX or Intel OSPray.

  • Radeon Pro Software for Enterprise 19.Q4 for Linux Released

    AMD on Tuesday released their Radeon Pro Software for Enterprise 19.Q4 for Linux package as their newest quarterly driver release intended for their professional graphics card offerings.

    Radeon Pro Software for Enterprise 19.Q4 for Linux is arriving as scheduled and continues to provide both the AMDGPU-PRO and AMDGPU-Open driver stacks depending upon your preferences.

Benchmarks Of JCC Erratum: A New Intel CPU Bug With Performance Implications On Skylake Through Cascade Lake

Filed under
Graphics/Benchmarks

Intel is today making public the Jump Conditional Code (JCC) erratum. This is a bug involving the CPU's Decoded ICache where on Skylake and derived CPUs where unpredictable behavior could happen when jump instructions cross cache lines. Unfortunately addressing this error in software comes with a performance penalty but ultimately Intel engineers are working to offset that through a toolchain update. Here are the exclusive benchmarks out today of the JCC erratum performance impact as well as when trying to recover that performance through the updated GNU Assembler.

The microcode update prevents jump instructions from being cached in the Decoded Icache when those instructions cross a 32-byte boundary or where they end on a 32-bit boundary. Due to that change there will be more misses from the Decoded ICache and switches back to the legacy decode pipeline -- resulting in a new performance penalty. The Decoded ICache / Decoded Streaming Buffer has been around since Sandy Bridge but only Skylake and newer is affected by this erratum. Cascade Lake is affected by this erratum but Ice Lake and future iterations appears unaffected. The erratum notice officially lists Amber Lake, Cascade Lake, Coffee Lake, Comet Lake, Kaby Lake, Skylake, and Whiskey Lake as affected generations for the JCC bug.

Read more

Linux driver patches indicate AMD is readying integer scaling

Filed under
Graphics/Benchmarks
Linux

Both Intel and Nvidia have released graphics driver updates to enable integer scaling options this year. Intel made a big song and dance out of the development process with Tweets and blog updates trailing the graphics driver feature. Then integer scaling became available for Intel Gen11 graphics users after a September driver update. Nvidia actually pipped Intel to the post by implementing integer scaling (for Turing GPUs) in its Gamescom driver release in August - it snuck in the update without much fanfare as it simultaneously boosted a number of AAA games performance and added some new image sharpening features.

Read more

Windows 10 vs. Ubuntu 19.10 vs. Clear Linux Performance On The Dell Ice Lake Laptop

Filed under
GNU
Graphics/Benchmarks
Linux

Last month I posted benchmarks looking at the Windows 10 vs. Linux OpenGL and Vulkan graphics performance for the Ice Lake "Gen11" graphics. But for those wondering about the CPU/system performance between Windows and Linux for the Core i7-1065G7 with the Dell XPS 7390, here are those benchmarks as we compare the latest Windows 10 to Ubuntu 19.10 and Intel's own Clear Linux platform.

Read more

Also: Virtual KMS Driver To Work On Virtual Refresh Rate Support (FreeSync)

AMD/Radeon: Driver for Linux 5.5, Radeon Software for Linux 19.30, and Benchmarks

Filed under
Graphics/Benchmarks
Linux
  • AMD Volleys Another Batch Of Graphics Driver Changes For Linux 5.5

    While the Linux 5.4 cycle is quickly winding down and with DRM-Next's cut-off crossing, AMD has sent in a last minute batch of changes it's targeting for the upcoming Linux 5.5 merge window.

    AMD in prior weeks submitted a lot of new GPU driver code for Linux 5.5. Friday's pull request is primarily fixes but one notable addition is enabling dynamic power gating for GCN with Raven Ridge APUs.

  • Radeon Software For Linux Updated With Radeon RX 5500 Series Support

    Radeon Software for Linux 19.30 has been the driver release branch since July for the AMD Linux packaged driver stack. That 19.30 driver was introduced with the AMD Radeon RX 5700 "Navi" support while now a slightly updated stack was released.

    Released on Friday was the AMDGPU-PRO 19.30-934563 driver stack as the newest revision to the Radeon Software for Linux 19.30 driver.

  • Benchmarks Of 10 Higher-End Intel/AMD CPUs On Ubuntu 19.10

    With Ubuntu 19.10 bringing some CPU/system performance changes compared to earlier Ubuntu releases as a result of compiler/toolchain upgrades, the newer kernel, and more, here is a quick weekend look at how the Ubuntu 19.10 performance compares across ten different AMD Ryzen and Intel Core systems.

    This is a reference look at the Intel/AMD performance on ten different higher-end desktop/workstation systems with a variety of workloads on Ubuntu 19.10 given the package upgrades found in this recent Linux distribution release.

Games and Graphics: Mainframe Defenders, Proton, NVIDIA 390.132 Linux Driver and More

Filed under
Graphics/Benchmarks
Gaming
  • Mainframe Defenders, retrofuturistic squad-based strategy now has a Linux demo

    Mainframe Defenders from Old Byte Apps is an in-development retro take on a squad-based strategy roguelike. They just announced today that they've now added a Linux demo.

    The story is your typical cyberspace affair, with a virus infiltrating infecting an AI controlled research complex. Your overall mission is to crush this virus and defend the mainframe through a series of missions with different objectives and enemies.

  • Steam Play Proton 4.11-8 is out with vkd3d for Direct3D 12 support

    Another update to Steam Play Proton has been released this evening, which should bring with it plenty of improvements for playing Windows games on Linux.

    Looking to get started with Steam Play on Linux? Be sure to check our previous beginners guide.

    Proton 4.11-8 now includes vkd3d, another library built on top of Vulkan to add in Direct3D 12 support. Other parts of Proton also saw version bumps like DXVK to 1.4.4, D9VK to 0.30, FAudio to 19.11 and Wine-mono to 4.9.4.

  • NVIDIA 390.132 Linux Driver Released For Legacy Fermi Support

    Not nearly as exciting as the recent NVIDIA 440 Linux driver series going stable but for those with older Fermi graphics cards and wanting to use the latest NVIDIA binary driver experience, their 390 series legacy driver series has been updated.

    The NVIDIA 390.132 driver is out today as the latest legacy driver update targeting the GeForce GTX 400/500 "Fermi" graphics cards.

  • Your weekend look at what's on sale and what you can try free

    While the Halloween sales are over, plenty of stores still have some big and interesting game sales going on with lots of Linux games going cheap.

mesa 19.3.0-rc2

Filed under
Graphics/Benchmarks

Hi list,

Along with the stable mesa 19.2.3 release today, I'm pleased to announce mesa
19.3.0-rc2. There's a large number of bug fixes in this release, nouveau, intel,
radeon, radv, turnip, nir, meson, zink, iris, swr, core mesa, and android fixes
are all present here.

Developers, please have a look at the release tracker:
https://gitlab.freedesktop.org/mesa/mesa/-/milestones/5, there's plenty of bugs
that need looking at if you have some time.

Dylan


Shortlog
========

Alyssa Rosenzweig (1):
      pipe-loader: Build kmsro loader for with all kmsro targets

Bas Nieuwenhuizen (6):
      radv: Fix timeout handling in syncobj wait.
      radv: Remove _mesa_locale_init/fini calls.
      turnip: Remove _mesa_locale_init/fini calls.
      anv: Remove _mesa_locale_init/fini calls.
      radv: Fix disk_cache_get size argument.
      radv: Close all unnecessary fds in secure compile.

Daniel Schürmann (4):
      docs/relnotes/new_features.txt: Add note about ACO
      aco: fix immediate offset for spills if scratch is used
      aco: only use single-dword loads/stores for spilling
      aco: fix accidential reordering of instructions when scheduling

Dylan Baker (3):
      nir: correct use of identity check in python
      meson: Add dep_glvnd to egl deps when building with glvnd
      Bump VERSION to 19.3.0-rc2

Erik Faye-Lund (1):
      zink: emit line-width when using polygon line-mode

Ian Romanick (1):
      intel/compiler: Report the number of non-spill/fill SEND messages on vec4 too

Ilia Mirkin (2):
      gm107/ir: fix loading z offset for layered 3d image bindings
      nv50/ir: mark STORE destination inputs as used

Jan Zielinski (1):
      gallium/swr: Fix depth values for blit scenario

Jason Ekstrand (3):
      anv: Fix a potential BO handle leak
      anv/tests: Zero-initialize instances
      anv: Set the batch allocator for compute pipelines

Jordan Justen (2):
      iris: Add IRIS_DIRTY_RENDER_BUFFER state flag
      iris/gen11+: Move flush for render target change

Kenneth Graunke (1):
      iris: Fix "Force Zero RTA Index Enable" setting again

Lionel Landwerlin (3):
      intel/dev: set default num_eu_per_subslice on gen12
      mesa: check draw buffer completeness on glClearBufferfi/glClearBufferiv
      anv: Properly handle host query reset of performance queries

Mauro Rossi (1):
      android: aco: fix Lower to CSSA

Paulo Zanoni (1):
      intel/compiler: remove the operand restriction for src1 on GLK

Pierre-Eric Pelloux-Prayer (2):
      radeonsi: tell the shader disk cache what IR is used
      mesa: enable msaa in clear_with_quad if needed

Samuel Pitoiset (1):
      radv: fix compute pipeline keys when optimizations are disabled


git tag: mesa-19.3.0-rc2

Read more

Also: Mesa 19.3-RC2 Released With Fixes To RADV Vulkan, Intel Driver Fixes

Graphics: Mesa 19.2.3 and NVIDIA Xavier NX

Filed under
Graphics/Benchmarks
  • mesa 19.2.3
    Hi list,
    
    I'd like to announce the immediate availability of mesa 19.2.3. Things are
    mostly slowing down now, the one exception is the giant pile of release-script
    changes from me. Yay simplifying the release.
    
    We've got a bit of everything in this release, iris, meson, radv, anv, turnip ,
    965, svga, utils, core mesa, glsl, etanviv, and gallium/rbug. But not too much
    any one place, all in all it feels like we're settling nicely into the stable
    release groove.
    
    Dylan
    
    
    Shortlog
    ========
    
    
    Bas Nieuwenhuizen (4):
          radv: Fix timeout handling in syncobj wait.
          radv: Remove _mesa_locale_init/fini calls.
          turnip: Remove _mesa_locale_init/fini calls.
          anv: Remove _mesa_locale_init/fini calls.
    
    Caio Marcelo de Oliveira Filho (1):
          anv: Fix output of INTEL_DEBUG=bat for chained batches
    
    Danylo Piliaiev (1):
          glsl: Initialize all fields of ir_variable in constructor
    
    Dylan Baker (13):
          bin/gen_release_notes.py: fix conditional of bugfix
          bin/gen_release_notes.py: strip '#' from gitlab bugs
          bin/gen_release_notes.py: Return "None" if there are no new features
          bin/post_version.py: Pass version as an argument
          bin/post_version.py: white space fixes
          bin/post_release.py: Add .html to hrefs
          bin/gen_release_notes.py: html escape all external data
          bin/gen_release_notes.py: Add a warning if new features are introduced in a point release
          cherry-ignore: update for 19.2.3 cycle
          nir: correct use of identity check in python
          meson: Add dep_glvnd to egl deps when building with glvnd
          docs: add release notes for 19.2.3
          Bump version to 19.2.3
    
    Ilia Mirkin (1):
          nv50/ir: mark STORE destination inputs as used
    
    Illia Iorin (1):
          Revert "mesa/main: Fix multisample texture initialize"
    
    Jason Ekstrand (2):
          anv: Fix a potential BO handle leak
          anv/tests: Zero-initialize instances
    
    Jon Turney (2):
          rbug: Fix use of alloca() without #include "c99_alloca.h"
          Fix timespec_from_nsec test for 32-bit time_t
    
    Jonathan Marek (1):
          etnaviv: fix depth bias
    
    Kenneth Graunke (1):
          iris: Fix "Force Zero RTA Index Enable" setting again
    
    Lionel Landwerlin (2):
          anv: fix unwind of vkCreateDevice fail
          mesa: check draw buffer completeness on glClearBufferfi/glClearBufferiv
    
    Marek Olšák (1):
          util/u_queue: skip util_queue_finish if num_threads is 0
    
    Nanley Chery (5):
          anv: Properly allocate aux-tracking space for CCS_E
          intel/blorp: Disable depth testing for slow depth clears
          iris: Clear ::has_hiz when disabling aux
          iris: Don't leak the resource for unsupported modifier
          iris: Disallow incomplete resource creation
    
    Paulo Zanoni (1):
          intel/compiler: remove the operand restriction for src1 on GLK
    
    Pierre-Eric Pelloux-Prayer (1):
          mesa: enable msaa in clear_with_quad if needed
    
    Sagar Ghuge (1):
          intel/blorp: Assign correct view while clearing depth stencil
    
    Samuel Pitoiset (4):
          radv: do not create meta pipelines with 16 samples
          radv: do not emit rbplus if attachments are undefined
          radv/gfx10: fix 3D images
          radv: fix vkUpdateDescriptorSets with inline uniform blocks
    
    Tapani Pälli (1):
          i965: setup sized internalformat for MESA_FORMAT_R10G10B10A2_UNORM
    
    Thomas Hellstrom (2):
          svga: Fix banded DMA upload unmap
          winsys/svga: Limit the maximum DMA hardware buffer size
    
    
    git tag: mesa-19.2.3
    
  • Mesa 19.2.3 Released With Many Fixes While Waiting For Mesa 19.3 In A Few Weeks

    Mesa 19.2.3 has a number of RADV Vulkan driver fixes (including a fix for 3D images with GFX10/Navi), various fixes to the Intel ANV Vulkan driver, a few alterations to the maturing Intel Iris Gallium3D driver, and other random fixes throughout the 3D stack.

  • NVIDIA Launches Jetson Xavier NX As 70x45mm 10~15 Watt "AI Supercomputer"

    NVIDIA announced today the newest member of the Jetson family: the Xavier NX as "the world's smallest supercomputer" coming in at smaller than the size of a credit/debit card. This mini supercomputer can deliver 21 TOPS for modern AI workloads while consuming less than 10 Watts or optionally a higher-performance 15 Watt mode.

    The Jetson Xavier NX is powered by a low-power version of the Tegra Xavier SoC. The Jetson Xavier NX offers six NVIDIA Carmel ARMv8.2 cores, a 384-core Volta GPU with 48 Tensor cores, dual NVDLA engines, 8GB of LPDDR4x memory, 16GB eMMC, Gigabit Ethernet, USB 3.1, and other functionality all off a 70x45 mm PCB and running off a +5V line.

Syndicate content

More in Tux Machines