Language Selection

English French German Italian Portuguese Spanish

Graphics/Benchmarks

Older Broadwell Graphics Performance Is Looking Good With The New Intel Gallium3D OpenGL Linux Driver

Filed under
Graphics/Benchmarks

A majority of our benchmarking of Intel's new Gallium3D OpenGL open-source driver is done with various "Gen9" graphics hardware given its proliferation and not yet having any Icelake Gen11 graphics hardware for Linux benchmarking. But with the Iris Gallium3D going back to supporting Broadwell "Gen8" graphics, here is a fresh look at how that oldest supported Intel hardware is working for this new Linux open-source OpenGL driver compared to the current default "i965" Intel OpenGL driver too.

Last week I provided an extensive look at the current Intel Gallium3D driver performance with the common Gen9 graphics hardware and the performance (and overall stability) of this new driver is looking great. It's looking like Intel is still on track for enabling that driver by default in Mesa before the 19.3 release at the end of the calendar year. Following that testing I was curious about Broadwell so I fired up an old Lenovo ThinkPad X1 Carbon laptop.

Read more

Building A Linux HTPC / Storage Server With The SilverStone CS381

Filed under
Graphics/Benchmarks

SilverStone recently sent over their CS381 chassis that has proven to be quite a versatile micro-ATX enclosure that can accommodate up to twelve hard drives (eight of which are hot-swappable) all while coming in at just 400 x 225 x 316mm. The SilverStone CS381 could work quite well as a Linux HTPC / DIY Steam Linux gaming living room PC or SOHO file server system with its compact size while offering immense storage potential. Here's more on the SilverStone CS381 and our build with using a Ryzen 5 3400G that is playing well under Linux with an ASUS B450 motherboard.

It's been a while since last taking a look at any SilverStone enclosure, but with continuing to be impressed by their high-end cases over the years, it was exciting to look at the CS381 from their Case Storage Series. The key features of this case are offering support for up to twelve HDD/SSDs, up to a microATX motherboard, and other components while occupying just 30 liters of space. The case can be positioned in either a vertical or horizontal position depending upon the environment and eight of the drives being hot-swappable primes the case for interesting storage server options.

Read more

Graphics: DXVK, Nouveau

Filed under
Graphics/Benchmarks
  • DXVK 1.4 Released With Updates Against Direct3D 11.4, Other Improvements

    In time for any weekend gaming is the release of DXVK 1.4 as the latest big update to this Direct3D 11 over Vulkan implementation to boost the D3D11 Windows gaming performance with the likes of Wine and Valve's Steam Play (Proton).

    With DXVK 1.4 the Direct3D interfaces have been updated against D3D11.4, the latest D3D11 revision shipped by Windows 10 Build 1903. This update brings new API features but DXVK isn't yet supporting some of the optional features like tiled resources and conservative rasterization.

  • Nouveau Finally Lands SPIR-V Support As Part Of OpenCL Push

    Going back to December 2017 we've been tracking the Red Hat led effort on improving Nouveau's OpenCL compute support that involves adding NIR/SPIR-V support and improvements to the Clover Gallium3D state tracker. To much surprise, this morning the SPIR-V support for this open-source NVIDIA driver was merged for Mesa 19.3.

Noctua NH-L9a-AM4: A Very Low-Profile AMD Ryzen Cooler

Filed under
Graphics/Benchmarks

At just 37mm tall, the Noctua NH-L9a-AM4 is one of the shortest yet quite capable CPU heatsink fans we have seen yet for AMD Ryzen processors.

When looking for a heatsink with a small stature for an AMD APU mini PC build for HTPC / file storage use-cases (more on that build in the next day or two), the Noctua NH-L9a-AM4 fit the criteria and so I went with that given the success with the many Noctua heatsinks we have used over the years. For those potentially interested in the NH-L9a-AM4 for an AMD APU like the new Ryzen 5 3400G or for lower-end Ryzen CPUs, I ran some benchmarks with this cooler.

Read more

Graphics: AMD, GNOME Shell on Wayland and NVIDIA Nsight Graphics

Filed under
Graphics/Benchmarks
  • AMD Pushes Back 3rd Gen Threadripper & Ryzen 9 3950X Until November

    While the Ryzen 9 3950X and 3rd Gen Ryzen Threadripper processors were reportedly on track for launching in October with updates as of a few weeks ago, today AMD announced a slight delay in launching these new processors.

  • AMD have delayed the Ryzen 9 3950X and 3rd generation Threadripper until November

    Today, AMD sent out a brief statement about a delay in their 16 core Ryzen 9 3950X and the 3rd generation Threadripper.

    [...]

    So if you were looking to grab either, keep an eye out in November. Will share any more news when they send it about the expected date and pricing.

  • AMD Sends In Initial Batch Of Fixes To Linux 5.4 - Includes Dali Support

    While just yesterday the big DRM feature pull was sent in for Linux 5.4, AMD has also volleyed out their initial batch of fixes for this next version of the kernel.

    This new AMDGPU pull isn't strictly fixes but as anticipated does include the recently reported Dali APU support. Dali along with Renoir -- also newly-supported in Linux 5.4 -- are some of AMD's 2020 APUs. Dali will be targeting the lower-end of the spectrum it's expected for value mobile/embedded. From the driver code, Dali looks like a newer revved version of the current-gen Picasso APUs. Both Dali and Renoir are based on the Vega architecture.

  • Linux Plumbers Conference 2019, part 2

    Pain points and missing pieces with Wayland, or specifically GNOME Shell:

    GNOME Shell is slower
    Synergy doesn't work(?) - needs to be in the compositor
    With Nvidia proprietary driver, mutter and native Wayland clients get GPU acceleration but X clients don't
    No equivalent to ssh -X. Pipewire goes some way to the solution. The whole desktop can be remoted over RDP which can be tunnelled over SSH.
    No remote login protocol like XDMCP
    No Xvfb equivalent
    Various X utilities that grab hot-keys don't have equivalents for Wayland
    Not sure if all X's video acceleration features are implemented. Colour format conversion and hardware scaling are implemented.
    Pointer movement becomes sluggish after a while (maybe related to GC in GNOME Shell?)
    Performance, in general. GNOME Shell currently has to work as both a Wayland server and an X compositor, which limits the ability to optimise for Wayland.

  • NVIDIA's Nsight Graphics 2019.5 Released With Better Vulkan Coverage

    NVIDIA this week released Nsight Graphics 2019.5 as the newest feature update to their proprietary developer tool for graphics profiling and debugging across multiple APIs.

    The Nsight Graphics 2019.5 release brings support for more than a dozen new Vulkan extensions, a variety of user-interface improvements, compatibility enhancements, and better syntax highlighting.

AMD EPYC 7642 Benchmarks: The Rome 48 Core CPU That Easily Takes On Intel's Xeon Platinum 8280

Filed under
Graphics/Benchmarks

Since the AMD EPYC 7002 series "Rome" launch at the beginning of August, it's been known how AMD's top-end (aside from the newly-announced EPYC 7H12) EPYC 7742 easily outperforms the Intel Xeon Platinum 8280 in most real-world benchmarks. The EPYC 7742 not only outperforms the Xeon Platinum 8280 in raw performance but also at a significantly lower cost and it gets even better with the EPYC 7642. We have been testing the EPYC 7642 48-core processors and even there the performance is generally ahead of a Xeon Platinum 8280 while being about half the cost of that flagship non-AP Intel Xeon Scalable Cascadelake processor.

Complementing our recent EPYC 7302 and EPYC 7402 benchmarks, today we are focused on the EPYC 7642 as the Rome 48-core / 96-thread processor. This 48 core processor has a 2.3GHz base clock and 3.3GHz boost clock while having 256MB of L3 cache, eight DDR4-3200 memory channels, 128 PCIe 4.0 lanes, and other features in common with the EPYC 7742 and other Rome processors. The EPYC 7642 carries a 50MHz base clock speed advantage over the 64 core EPYC 7742 but a 100MHz lower boost clock speed as the principal differences aside from the core/thread count. Both of these CPUs carry a 225 Watt TDP.

Read more

Running The AMD "ABBA" Ryzen 3000 Boost Fix Under Linux With 140 Tests

Filed under
Graphics/Benchmarks

Last week AMD's AGESA "ABBA" update began shipping with a fix to how the boost clock frequencies are handled in hopes of better achieving the rated boost frequencies for Ryzen 3000 series processors. I've been running some tests of an updated ASUS BIOS with this adjusted boost clock behavior to see how it performs under Linux with a Ryzen 9 3900X processor.

The AGESA 1.0.0.3 ABBA update has an improved boost clock frequency algorithm along with changes to the idle state handling. This AGESA update should better position AMD Ryzen 3000 processors with the boost clock behavior expected by users with better hitting the maximum boost frequency and doing so more aggressively.

Read more

Graphics: Gallium3D, Mesa, RADV, ACO

Filed under
Graphics/Benchmarks
  • Intel's Gallium3D OpenGL Driver Taps Another Optimization - ~32% For GFXBench

    Intel's new OpenGL Linux driver, their Gallium3D-based "Iris" implementation that is aiming to be the default before year's end, continues making striking progress.

    Just this past week when testing the very latest Mesa code for this Intel Gallium3D driver I was quite impressed with it near universally being faster than their existing "i965" Mesa driver. For some OpenGL workloads, this Gallium3D driver is significantly faster than the driver it's set to replace for Broadwell "Gen 8" graphics and newer.

  • Mesa's Disk Cache Code Now Better Caters To 4+ Core Systems

    Most Linux gamers these days should be running at least quad-core systems so Mesa 19.3 has been updated to reflect that reality with the number of CPU threads used by their disk cache.

  • Performance-Boosting DFSM Support Flipped On & Off For RADV Vulkan Driver

    Back in July of last year the RADV Vulkan driver enabled primitive binning and DFSM for this open-source Radeon Vulkan driver. Well, it thought it enabled DFSM support and paired with the binning did yield a minor performance benefit at the time for Raven Ridge APUs. But now it turns out the DFSM support wasn't properly wired up and is now addressed but is currently introducing a performance regression.

    RADV developer Bas Nieuwenhuizen added the actual DFSM (Deterministic Finite State Machine) support and mirrors the behavior of the RadeonSI OpenGL driver. With the DFSM support he found that it doubles the fill-rate of one of his test samples from around 16 to 32 pixels/cycles for Raven Ridge.

  • The Valve-funded shader compiler 'ACO' is being queued up for inclusion in Mesa directly (updated: merged)

    Back in early July, Valve announced their work on a new AMD GPU shader compiler for Mesa named ACO and now they're trying to get it pulled into Mesa directly.

    Their main aims with ACO were to get the "best-possible code generation for game shaders, and fastest-possible compilation speed" and to replace the currently used shader compiler from the massive LLVM project. It has certainly seemed promising, improving both shader compile time resulting in less stuttering and so helping to improve overall FPS and smoothness in Linux games when played on supported AMD GPUs.

  • Valve's ACO Shader Compiler For The Mesa Radeon Vulkan Driver Just Landed

    It was just two days ago that Valve's performance-focused "ACO" shader compiler was submitted for review to be included in Mesa for the "RADV" Radeon Vulkan driver. Just minutes ago that new shader compiler back-end was merged for Mesa 19.3.

    ACO, short for the AMD COmpiler, is the effort led by Valve at creating a more performant and optimized shader compiler for the Radeon Linux graphics driver. Besides trying to generate the fastest shaders, ACO also aims to provide speedy shader compilation too, as an alternative to the AMDGPU LLVM shader compiler back-end. Initially ACO is for the RADV Vulkan driver but it may be brought to the RadeonSI OpenGL driver in the future. At the moment ACO is in good shape for Volcanic Islands through Vega while the Navi shader support is in primitive form.

CPU/GPU/Graphics Stack: AMD EPYC, NVIDIA and Mesa Radeon Vulkan Driver

Filed under
Graphics/Benchmarks
  • AMD EPYC 7H12 Announced As New 280 Watt Processor For High Performance Computing

    From Rome, Italy this afternoon AMD not only announced more than 100 world records have been broken with their new EPYC "Rome" processors, but there is also a new SKU! Meet the EPYC 7H12.

    The EPYC 7H12 doesn't quite follow the naming convention of the rest of the EPYC Rome line-up announced back in August as it's a special part. The EPYC 7H12 is more akin to Intel's Cascadelake-AP line-up but with more broad availability and just a higher clocked / higher power part as opposed to tacking on extra dies. But it carries the same focus on delivering maximum HPC performance.

  • Nvidia Open Sources Its Deep Learning Compiler

    System architects and software teams now have the complete source for the fully open software and hardware inference platform.

  • NVIDIA Bringing Up Open-Source Volta GPU Support For Their Xavier SoC

    While NVIDIA doesn't contribute much open-source Linux driver code as it concerns their desktop GPUs (though they have been ramping up documentation), when it comes to Tegra/embedded is where they have contributed improvements and new hardware support to Nouveau and associated driver code in the past several years. NVIDIA's open-source Tegra/embedded contributions come as a result of customer demand/requirements. Their latest work is preparing to finally bring-up the "GV11B" Volta graphics found within last year's Tegra Xavier SoC.

  • Valve's ACO Shader Compiler Under Review For The Mesa Radeon Vulkan Driver

    The RADV "ACO" shader compiler announced by Valve back in July for the fastest compilation speeds and best possible code generation may soon be hitting mainline Mesa for the open-source AMD Linux graphics stack.

    The ACO shader compiler as an alternative to the existing AMDGPU LLVM shader compiler back-end has shown quite promising results for Linux games. ACO has become more featureful over time and is now largely at feature parity to the existing shader compilation support while generally offering some performance advantages, thanks to the effort and funding by Valve.

FreeBSD 12 & DragonFlyBSD 5.6 Running Well On The AMD Ryzen 7 3700X + MSI X570 GODLIKE

Filed under
Graphics/Benchmarks

For those wondering how well FreeBSD and DragonFlyBSD are handling AMD's new Ryzen 3000 series desktop processors, here are some benchmarks on a Ryzen 7 3700X with MSI MEG X570 GODLIKE where both of these popular BSD operating systems were working out-of-the-box. For some fun mid-week benchmarking, here are those results of FreeBSD 12.0 and DragonFlyBSD 5.6.2 up against openSUSE Tumbleweed and Ubuntu 19.04.

Back in July I looked at FreeBSD 12 on the Ryzen 9 3900X but at that time at least DragonFlyBSD had troubles booting on that system. When trying out the Ryzen 7 3700X + MSI GODLIKE X570 motherboard on the latest BIOS, everything "just worked" without any compatibility issues for either of these BSDs.

Read more

Syndicate content

More in Tux Machines

Lubuntu, A Once Great Distro, Is Falling Behind

Lubuntu used to be that Linux distribution that you referred a friend to in case he wanted a very lightweight, newbie-friendly yet elegant alternative for Windows. Up to its 18.04LTS release, it indeed worked as expected, but starting with 18.10 where the development team switched to using the Qt-based desktop LXQt instead of traditional LXDE, things started to break. As a short background, you should know that there was a desktop environment called “Razor-Qt”, which was a newly developed desktop based on the Qt toolkit that aimed to be lightweight and modern in the same time. There was also another team working on a Qt branch of LXDE (which is GTK-based) called LXDE-Qt. After a lot of discussions, both teams combined efforts and started to work on one project called LXQt. LXDE desktop is still working today, and is considered to be feature complete. But it was not even ported to GTK 3 like other desktops such as MATE and XFCE, instead, it’s still using the legacy GTK 2. Read more

Fedora Workstation 31 – What's new

Fedora has been leading the migration to Wayland since day one and we are not planning to stop. XWayland on demand has been an effort a lot of people contributed to this cycle. The goal is to only need XWayland for legacy X applications, not have it started and running all the time as that is a waste of system resources and also having core functionality still depend on X under Wayland makes the system more fragile. XWayland-on-demand has been a big effort with contributions from a lot of people and companies. One piece of this was the Systemd user session patches that was originally written by Iain Lane from Canonical. They had been lingering for a bit so Benjamin Berg took those patches on for this cycle and helped shepherd them over the finish line and get them merged upstream. This work wasn’t a hard requirement for Wayland-on-demand, but since it makes it a lot easier to do different things under X and Wayland which in turn makes moving towards XWayland-on-demand a little simpler to implement. That work will also allow (in future releases) us to do things like only start services under GNOME that are actually needed for your hardware, so for instance if you don’t have a bluetooth adapter in your computer there is no reason to run the bits of GNOME dealing with bluetooth. So expect further resource savings coming from this work over time. Carlos Garnacho then spent time going through GNOME Shell removing any lingering X dependencies while Olivier Fourdan worked on cleaning up the control center. This work has mostly landed, but it is hidden behind an experimental flag (gsettings set org.gnome.mutter experimental-features "[...,'autostart-xwayland']") in Fedora 31 as we need to mature it a bit more before its ready for primetime. But we hope and expect to have it running by default in Fedora Workstation 32. Read more Also: Fedora Workstation 31 Should Be Another Fantastic Release For Desktop Linux

today's leftovers

  • Lima Gallium3D Picks Up A Buffer Object Cache, Partial Updates

    The Lima Gallium3D driver for supporting Arm Mali 400/450 graphics hardware within Arm SoCs has picked up a few performance optimizations. Vasily Khoruzhick has contributed a buffer object cache to this Gallium3D driver to avoid the great overhead costs to allocating buffer objects. The BO cache for Lima is modeled after the Broadcom V3D Gallium3D driver's BO cache.

  • Akademy Behind!

    The framework-for-that idea lives on, though: today I was looking for something to extract the Exec= line from a .desktop file, and there’s a framework for that (KIO does the job, but that’s a pretty heavy dependency for what I wanted; I’ll need to think about it some). Third year running (Almeria, Vienna, Milan), I presented the BoF wrap-up session at the end of the day – that’s mostly acting as MC to get other people to tell their stories. Here’s thursday and friday for instance, via the dot. Videos are on YouTube. For next year, I’d like to train some other people to do the presentation – because there are so many other faces in KDE. I have high hopes for Caio (of KPMCore and other things) and Aish (GCompris) who I’d like to see out there fronting for the KDE community. For being a loud person I’ve now been appropriately punished, by being voted on to the board of KDE e.V. (that page needs an update). I’ll be doing generally useful things, I hope, which means massaging the community code of conduct and spending money from our donors on events all over the world where people from all corners of the KDE community can participate.

  • ClonOS 19.09-RELEASE

    FreeBSD 13.0-CURRENT (r352386) cloud-init support (see errata) fixed fbbuf/vesa video issue with some recent Linux distros ( Kali Linux, Parrot, etc..) p9fs support (*) CBSD updated to 12.1.1 fixed known SQL injection vulnerabilities

  • EndeavourOS 2019.09.15 overview | AN ARCH-BASED DISTRO WITH A FRIENDLY COMMUNITY IN ITS CORE

    In this video, I am going to show an overview of EndeavourOS 2019.09.15 and some of the applications pre-installed.

  • Cameron Kaiser: A quick note for 64-bit PowerPC Firefox builders

    If you build Firefox on 64-bit Linux, *BSD, etc. for your G5, you may want to check out this Talospace article for an upcoming low-level fix especially as we need to ensure big-endian systems work fine with it. The problem never affected OS X Firefox for Power Macs because those builds were only ever 32-bit, and even TenFourFox is 32-bit through and through even on the G5 largely for reasons of Carbon compatibility which we need for some pieces of the widget code. Since this is syndicated on Planet Mozilla let me give a big thanks to Ted Campbell for figuring out the root cause, which turned out to be a long-standing problem I don't think anyone ever noticed before.

  • Preventing Lubuntu 18.04 from leaving a user process running after the user logs out
  • List Device Names, Disk and Partition Information in Linux with lsblk

Programming Leftovers

  • HEURISTIC ALGORITHMS FOR THE PROBLEM OF OPTIMIZATION OF THE DRAIN AREA IN UNCONVENTIONAL FIELDS OF HYDROCARBONS

    In a previous work, a linear programming based procedure was presented for this problem, and this procedure allowed to solve real-life instances, albeit -in some cases- producing plans with overlappings among the pads (since a relaxation is solved) and suffering from memory issues.

  • Getting Started With Async Features in Python

    Have you heard of asynchronous programming in Python? Are you curious to know more about Python async features and how you can use them in your work? Perhaps you’ve even tried to write threaded programs and run into some issues. If you’re looking to understand how to use Python async features, then you’ve come to the right place.

  • Anaconda Enterprise Receives Honors in Fourth Annual Datanami Readers’ and Editors’ Choice Awards

    Anaconda’s enterprise data science platform has been recognized in the fourth annual Datanami Readers’ and Editors’ Choice Awards, presented during the Strata Data Conference.

  • Exploratory Data Analysis Made Easy At The Command Line

    There are countless tools and libraries in Python for data scientists to perform powerful analyses, but they often have a setup cost that acts as a barrier to ad-hoc exploration of data. Visidata is a command line application that eliminates the friction involved with starting the discovery process. In this episode Saul Pwanson explains his motivation for creating it, why a terminal environment is a useful place for this work, and how you can use Visidata for your own work. If you have ever avoided looking at a data set because you couldn't be bothered with the boilerplate for a Jupyter notebook, then Visidata is the perfect addition to your toolbox. [...] There are countless tools and libraries in Python for data scientists to perform powerful analyses, but they often have a setup cost that acts as a barrier to ad-hoc exploration of data. Visidata is a command line application that eliminates the friction involved with starting the discovery process. In this episode Saul Pwanson explains his motivation for creating it, why a terminal environment is a useful place for this work, and how you can use Visidata for your own work. If you have ever avoided looking at a data set because you couldn’t be bothered with the boilerplate for a Jupyter notebook, then Visidata is the perfect addition to your toolbox.