Language Selection

English French German Italian Portuguese Spanish

openSUSE 10.3 in review: A solid Linux desktop

Filed under
Reviews

The openSUSE team officially released version 10.3 on October 4th. openSUSE is a popular German Linux distribution that Distrowatch.com lists as one of the "top ten." Underneath its new green artwork, version 10.3's improvements over previous versions include cutting down the time it takes to reach the graphical login screen; speeding up and streamlining its package management utility; and making it easier for users to install software using a new "one-click install" process.


About SuSE

SuSE began in 1992 as a German consulting group that distributed a German version of Slackware. Eventually they changed their distro to be RPM-based, and added an easy-to-use installation and administration tool named YaST ("Yet another Setup Tool"). However, YaST was originally proprietary software. Desktop verions of SuSE Linux were traditionally sold in boxed sets that included excellent written documentation. (I have a SuSE Linux 7.3 boxed set on the shelf that includes no less than four manuals and one "quick install guide.")

Users who didn't want to purchase SuSE originally had to wait a few months before a version was released that could be downloaded and installed via FTP. That all changed after Novell bought SuSE in 2004. With version 9.2, SuSE released a DVD version for public download, and, with version 10, opened the distro up for public participation in development and beta testing. They also released YaST under the GPL. They renamed the project "openSUSE" to reflect this change. (You can, of course, still purchase a boxed version of openSUSE, and install it via FTP, should you so desire.)

The aftermath of Novell's purchase of SuSE hasn't been a bed of roses. In late 2006, Novell and Microsoft announced an agreement to collaborate on interoperability between Novell's "Enterprise Linux" products and Microsoft products. Part of that agreement included a promise from Microsoft not to sue users of the Novell products covered in the agreement (including portions of openSUSE), for what Microsoft claims is patent infringement.

In short, Microsoft claims Linux infringes on hundreds of its software patents. By making this deal with Microsoft, Novell (at least implicitly) agreed with that position. This obviously did not go over well with many Linux users.

(For more information, see Groklaw's "Novell-MS Deal" page. For more information on the history of SuSE, see its Wikipedia entry.)


From the "Haste Makes Waste" Dept.

openSUSE doesn't have a live installation CD yet (like PCLinuxOS and Ubuntu do), although word is that one's in the works. It uses the traditional method of booting off the install media, with YaST guiding you through the installation process. Installation is a two-part process. The first part involves partitioning; boot loader configuration and installation; and package selection and installation. Once that's done, the installer reboots your computer, and restarts itself from your hard drive. The second half of the process involves configuring the root password; installing any available online updates (which, once again, requires you to configure your network connection -- this time for good); configuring user accounts; and hardware configuration.

There are a few different ways one can install openSUSE 10.3. On my test box (a $150 Fry's special based on an AMD Sempron 2200+ CPU, which now has 640MB of RAM, a NVIDIA GeForce MX 440 video card, and a larger hard drive), I used the "KDE-only" and the "non-OSS add-on" CDs. (A "GNOME-only" installation CD is also available.) The advantages of having KDE and GNOME installation CDs is that they take a lot less time to download than the full installation DVD, and why bother if you're never going to install the other desktop environment? (For fans of Xfce, you can also install Xfce 4.4.1. Also, if you want to beta-test the next version of KDE, you can install KDE 4 alongside KDE 3.)

On my Compaq Presario V2610US laptop, I used the full installation DVD.

In order to use online repositories in addition to the installation media, the installer gives you the option of configuring your Internet connection at about the third step in the process. In other words, if you use the KDE or GNOME installation CDs and have a slow Internet connection or a download cap, be forewarned that the installer will go out and grab packages from the Internet if you tell it that it can. (It probably does the same thing using the DVD, but my laptop wasn't physically plugged in to the network, so I skipped that step.)

   

L: Installation using the KDE and Non-OSS add-on CDs. Note that the installer's pulling most of the packages off the Internet.
R: A successful installation


Overall, I had a rather odd installation experience. I was in a rush during my first attempt to install openSUSE on my Fry's box, trying to squeeze as much as possible into a spare hour. I somehow managed to get into dependency resolution hell. How does one get a dependency error by selecting a package in an installer? one wonders. It went downhill from there. Long story short, you can get into a mess if you're in a hurry. Take your time. Or simply accept the default package selection and make changes after the installation's finished, which is what I ended up doing when I restarted later.

On my laptop, the second package YaST tried to install failed to pass its integrity check. I had a bad burn. Fortunately, I had the DVD ISO image on my workstation, and burned a new one at 4x. From there, it was fairly easy to go to a virtual console on the laptop, look at /proc/mounts to find out where the DVD was mounted, unmount it, eject it, and mount the newly-burned DVD to the same directory. Afterwards, installation finished with no problems.

A new installation of openSUSE 10.3 installs:

  • Linux kernel 2.6.22.5-31
  • X.org 7.2
  • KDE 3.5.7 (with a few games from the upcoming KDE 4 thrown in)
  • GNOME 2.20
  • Firefox 2.0.0.6
  • Thunderbird 2.0.0.6
  • GIMP 2.2.17
  • OpenOffice.org 2.3
  • K3b 1.0.3

(After a week of online updates, among other things, the kernel version's currently 2.6.22.9-0.4.) For a comprehensive list of features, see this "Product Highlights" page.

   

Games from KDE 4; the "sysinfo" kio slave (a.k.a. "My Computer")


Both my Fry's test box and my laptop are "special needs" cases in terms of proprietary video drivers. The Fry's box's GeForce MX 440 likes version 1.0.9631 of the NVIDIA driver. The laptop, with its ATI Radeon Mobility 200M chipset, likes ATI driver 8.40.4. This, of course, means manual installation, rather than the much easier one-click installation method for NVIDIA (the same page tells you how to install the driver manually) or the corresponding one-click installation method for ATI. (I used the instructions in this blog post to install the ATI driver. It's aimed at users of openSUSE 10.2, but the 8.40.4 ATI driver doesn't recognize openSUSE 10.3 as an option anyway.)

Some other things of note:

  • Both my test box and my laptop dual-boot with Windows. openSUSE set up ntfs-3g with read/write support for my NTFS partitions automatically.
  • openSUSE now includes MP3 support out of the box.
  • My laptop's Broadcom wireless card requires the use of ndiswrapper and the Windows drivers. (That link gives comprehensive instructions on how to install and use ndiswrapper on openSUSE.) For some reason, recent kernels include an open-source "bcm43xx" module that doesn't work with my laptop's BCM4318 chipset (one would assume they work with some Broadcom chipset out there), but the kernel's hotplugging system keeps loading it anyway. In some other versions of Linux I've tried, it interfered with ndiswrapper, which meant the "bcm43xx" module had to be blacklisted. In openSUSE 10.3 (using ndiswrapper and KNetworkManager), things seem to work fine without blacklisting it.
  • The openSUSE wiki has a good two-part article on getting suspend-to-RAM and suspend-to-disk working. (For it to work on my laptop, which doesn't have an ACPI-aware BIOS, it means using the "radeon" driver rather than the accellerated "fglrx" driver, and turning off the framebuffer and booting in 80x25 character mode. So the trade-off is a) no 3D desktop effects and Cool an ugly-looking boot sequence.)


One-Click Install and Other Improvements

openSUSE 10.3 seems faster than openSUSE 10.2 in three major ways. First, you get to a login prompt more quickly. openSUSE seems to do this by starting the display manager before it's done loading every startup service (Debian "Lenny" does the same thing). It's also made its package manager feel faster than it did before, by taking less time to start up. Finally, they've removed the ZENworks Management Daemon (ZMD), which caused a noticeable slowdown in openSUSE 10.2, especially when running at the same time as the Beagle indexing service.


The openSUSE Updater in action


YaST now includes a "Community Repositories" applet that checks for available online software repositories and lets you add them with a click of the mouse. (Of course, the more repositories you add, the slower YaST's package manager will be!) You can also add repositories manually, using the "Software Repositories" applet.


Adding community repositories


Perhaps the feature that'll garner the most attention is openSUSE's one-click install. This involves opening an XML file (usually from a web page) that has the extention "*.ymp" (presumably for "YaST meta-package") with the YaST Meta Package Handler. For example, this page on the openSUSE Community website includes *.ymp files for both KDE and GNOME users that allows you to install a wide variety of proprietary codecs (and a selection of applications) simply by clicking the links, letting YaST open them, and following the prompts. This replaces the old method of manually adding repositories, then searching for and installing the software you want (although you can still do it that way, if you want to). It also has security implications, in that you could install malicious software by mistake; there's a dialog box that comes up during the process, warning you of that fact.



Using one-click install — well, almost.


You can also install compiz-fusion (the successor to Beryl) through a one-click install process (including a "fusion-icon" applet that lives in the system tray, similar to beryl-manager). While that article makes it seem like you need to run XGL in order to use compiz-fusion, that's certainly not true for those who have NVIDIA cards. Those folks will simply need to follow the same steps (see step #4) they used to enable compositing in /etc/X11/xorg.conf that they used with Beryl.

   


Fun with compiz-fusion


In Conclusion...

openSUSE 10.3 is a mature, stable, free Linux distribution. Pros include excellent documentation and a wide selection of up-to-date software. Its YaST configuration utility offers ease-of-use matched perhaps only by Mandriva's Control Center. It seems as fast and responsive as any other modern Linux distro, even on my rather underpowered test box (except for the 3D desktop effects, which is to be expected).

Cons include its clunky software management tool, which still takes too much time to load, despite efforts to improve it. (In the grand scheme of things, however, this seems like a small complaint to make.) And, while it has little to do with openSUSE's development team, a big raspberry goes to Novell's marketing geniuses for giving credence to Microsoft's patent claims in exchange for some amount of money.


openSUSE Resources

  • For a blue version of the default desktop wallpaper, check here


openSUSE in blue



Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.

Nice

Hey, thanks for the nice review Smile

I've installed OpenSuSE 10.3 recently, but I didn't have a green colourscheme applied to the window decorations by default (I chose KDE desktop off the DVD install). However, I notice you do in your screenshots! Did you do that yourself or was it automatic for you?

Thanks.

yes

Yes, he did that himself. Mine was also defaulted to blue.

That's correct, it's changed from the default

The screenshot at the top is the only one of the default desktop. The rest of the "green machine" screenshots have QTCurve as the style, a custom wallpaper, and a custom green color scheme. (If you don't like the way KDE looks in whatever distro you have, change it!)

If you like the color scheme, save the following to a file named "green.kcsrc" and import it in kcontrol's "colors" module:

[Color Scheme]
Name=green
activeBackground=65,127,75
activeBlend=114,174,114
activeForeground=255,255,255
activeTitleBtnBg=220,220,220
alternateBackground=240,240,240
background=238,238,230
buttonBackground=238,234,222
buttonForeground=0,0,0
contrast=7
foreground=0,0,0
frame=238,238,230
handle=238,238,230
inactiveBackground=116,132,115
inactiveBlend=226,230,199
inactiveForeground=255,255,255
inactiveFrame=238,238,230
inactiveHandle=238,238,230
inactiveTitleBtnBg=220,220,220
linkColor=0,0,192
selectBackground=134,204,134
selectForeground=0,0,0
shadeSortColumn=true
visitedLinkColor=128,0,128
windowBackground=255,255,255
windowForeground=0,0,0

More in Tux Machines

Security: Updates, VLC FUD and LinuxSecurity Turning 20

  • Security updates for Tuesday

    Security updates have been issued by Debian (libsdl2-image and libxslt), Oracle (java-1.8.0-openjdk and java-11-openjdk), Scientific Linux (java-1.8.0-openjdk and java-11-openjdk), SUSE (bzip2, microcode_ctl, and ucode-intel), and Ubuntu (clamav, evince, linux-hwe, linux-gcp, linux-snapdragon, and squid3).

  • Dodgy vids can hijack PCs via VLC security flaw, US, Germany warn. Software's makers not app-y with that claim

    In a bug-tracking ticket discussing CVE-2019-13615, VideoLAN lead developer Jean-Baptiste Kempf noted that he was unable to recreate the crash using a proof-of-concept .MP4 video, provided by a security researcher four weeks ago, that's supposed to knacker the latest version of VLC, 3.0.7.1. Nor was he able to crash the older 3.0.6 and work-in-progress releases, such as 3.0.8, he reported. "This does not crash a normal release of VLC 3.0.7.1," added Kempf. "Sorry, but this bug is not reproducible and does not crash VLC at all." VLC developer Francois Cartegnie was more blunt earlier today: "If you land on this ticket through a news article claiming a critical flaw in VLC, I suggest you to read the above comment first and reconsider your (fake) news sources."

  • Our Linux Sister Linuxsecurity.com are Celebrating their 20th Anniversary by Launching a New Website

    LinuxSecurity.com is the community’s central source for information on Linux and open source security. They follow the open source trends as they affect the community. Also they produce content that appeals to administrators, developers, home users, and security professionals. Having created a site that satisfies the needs of both IT professionals – including engineers, programmers, designers and system administrators – and those individuals seeking to learn more about security and open source, LinuxSecurity.com has grown to encompass not only their website but also two leading industry email newsletters, Linux Security Newsletter and Security Advisories Weekly.

Fedora and IBM/Red Hat Leftovers

  • ThinkPad X1 Carbon 6th gen on Fedora
  • Multitenant deployment of MongoDB using OpenShift Container Storage and using YCSB to test performance
  • IBM gives cancer-killing drug AI project to the open source community

    IBM has released three artificial intelligence (AI) projects tailored to take on the challenge of curing cancer to the open-source community. At the 18th European Conference on Computational Biology (ECCB) and the 27th Conference on Intelligent Systems for Molecular Biology (ISMB), which will be held in Switzerland later this month, the tech giant will dive into how each of the projects can advance our understanding of cancers and their treatment. 

  • IBM Open Sources Cancer-Fighting AI Project

    Now, the company has decided to make all three tools open-source, meaning scientists will be able to use them in their research whenever they please, according to ZDNet. The tools are designed to streamline the cancer drug development process and help scientists stay on top of newly-published research — so, if they prove useful, it could mean more cancer treatments coming through the pipeline more rapidly than before.

  • An OpenShift Administrator’s Guide to Onboarding Applications

    Infrastructure teams managing Red Hat OpenShift often ask me how to effectively onboard applications in production. OpenShift embeds many functionalities in a single product and it is fair to imagine an OpenShift administrator struggling to figure out what sort of conversations his team must have with an application team before successfully running an application on OpenShift. In this article, I suggest a few topics that administrators could use to actively engage with fellow application teams for the onboarding process. I have had several conversations with customers on these topics and observed that suggested approach has really helped them. By no means are these topics exhaustive, but they are sufficient to kick start the necessary and relevant conversations. Over time, I expect administrators to have larger conversations with application teams in application onboarding. 

  • OpenWhisk Gets Its Apache Software Diploma

    The OpenWhisk open source serverless platform hit graduation status as a Top-Level Project at the Apache Software Foundation. The designation comes as the serverless ecosystem continues its rapid evolution in meeting the production needs of organizations. The OpenWhisk project itself was initially born out of IBM, which donated its beta-level code into the Apache Incubator project in late 2016. IBM was using that codebase to support functions running on its IBM Cloud.

  • The browser wars and the birth of JavaScript

    Before anything like an Android device or iPhone existed, desktop computers were the battleground for the browser wars. The battle involved billions of dollars invested by a number of companies, all based on the premise that whoever ruled the desktop browser market would own the internet. Today, mobile devices account for nearly half of all website traffic. Back in the 1990s, however, almost all of the action on the web came from desktop machines, and the vast majority of those desktop machines were running some flavor of Microsoft Windows. In the browser world, the first-mover advantage belonged to Netscape Communications Corporation. They built the Netscape Navigator browser that made the web accessible to millions for the first time. Netscape had more than 80% of the market, but they also had no shortage of competition. IBM had a browser for OS/2. Oracle had the Powerbrowser, a Netscape-compatible product that included something called the Database Markup Language. The real danger to Netscape, of course, came from the company that owned more than 80% of the world’s desktops: Microsoft. Strategically, Netscape realized that the web needed to move past static web pages to reach its full potential. Even if they were created dynamically by something like a CGI script on the web server, pages didn’t change once they arrived in your browser. If you wanted to see even a slightly modified version of a page, you had to send a request back to the server and wait for a response. For all its sophistication, a web browser felt a lot like a dumb terminal attached to a mainframe. What web developers needed was a programming language that would run in the browser, taking advantage of the processing power of the desktop machine to give users a richer experience. [...] JavaScript’s dominance was cemented by the emergence of Node.js on the server side. At a minimum, it gave web developers the ability to take their JavaScript skills from the client to the server. Combined with the ability to pass functions as objects (callbacks), Node.js’s event loop popularized a whole new programming model. Suddenly, you could write a web server in just a few lines of code. Then, the rise of the Node Package Manager (npm) to manage dependencies meant a very small application could leverage other packages to do really sophisticated things. As a self-serving example, the knative-proxy package in the Coderland Compile Driver needs fewer than 40 lines of code to handle the HTTP POST and OPTIONS verbs. And it took yr author maybe 30 minutes to write. JavaScript is a simple, unpretentious language that has its fingers in every corner of your life. Turn off JavaScript in your browser and see how much of the web doesn’t work anymore. (Philosophical arguments as to whether that’s a good or bad thing are left to the reader.) No matter how or where you use the internet, Brendan Eich’s 10-day coding spree is the most important sprint in the history of computing. You don’t have to like JavaScript, but if you make a living developing for the web, you have to learn it.

Programming: Python, GCC and More

  • Stack Abuse: Python for NLP: Word Embeddings for Deep Learning in Keras

    This is the 16th article in my series of articles on Python for NLP. In my previous article I explained how N-Grams technique can be used to develop a simple automatic text filler in Python. N-Gram model is basically a way to convert text data into numeric form so that it can be used by statisitcal algorithms. Before N-Grams, I explained the bag of words and TF-IDF approaches, which can also be used to generate numeric feature vectors from text data. Till now we have been using machine learning appraoches to perform different NLP tasks such as text classification, topic modeling, sentimental analysis, text summarization, etc. In this article we will start our discussion about deep learning techniques for NLP. Deep learning approaches consist of different types of densely connected neural networks. These approaches have been proven efficient to solve several complex tasks such as self-driving cars, image generation, image segmentation, etc. Deep learning approaches have also been proven quite efficient for NLP tasks. In this article, we will study word embeddings for NLP tasks that involve deep learning. We will see how word embeddings can be used to perform simple classification task using deep neural network in Python's Keras Library.

  • Python with JSON Files

    With the growth and evolution of challenges in computer science, Python continues to rise as the primarily sought-after programming skill to solve data science problems.

  • Logging in Python

    Logging is a very useful tool in a programmer’s toolbox. It can help you develop a better understanding of the flow of a program and discover scenarios that you might not even have thought of while developing. Logs provide developers with an extra set of eyes that are constantly looking at the flow that an application is going through. They can store information, like which user or IP accessed the application. If an error occurs, then they can provide more insights than a stack trace by telling you what the state of the program was before it arrived at the line of code where the error occurred.

  • Let’s Build A Simple Interpreter. Part 16: Recognizing Procedure Calls

    Today we’re going to extend our interpreter to recognize procedure calls. I hope by now you’ve flexed your coding muscles and are ready to tackle this step. This is a necessary step for us before we can learn how to execute procedure calls, which will be a topic that we will cover in great detail in future articles. The goal for today is to make sure that when our interpreter reads a program with a procedure call, the parser constructs an Abstract Syntax Tree (AST) with a new tree node for the procedure call, and the semantic analyzer and the interpreter don’t throw any errors when walking the AST.

  • Playing Tic Tac Toe using Reinforcement Learning

    I have always been fascinated by the amazing work done by OpenAI. The one that stood out to me was a AI bot that could play the massively popular game - Dota 2. Dota 2 used to be the escape from the real world for me and my friends while I was in high school. This inspired me to learn more about the field of RL. I wanted to start small so I started with Tic Tac Toe.

  • Python zip function tutorial (Simple Examples)

    The zip() function in Python programming is a built-in standard function that takes multiple iterables or containers as parameters. An iterable in Python is an object that can be iterated or stepped through like a collection. The zip() function is used to map the same indexes of more than one iterable. Mapping these indexes will generate a zip object.

  • GCC 10 Compiler Picks Up New Scheduler Model & Cost Tables For AMD Zen 2 Processors

    While AMD developers published their "Znver2" compiler patches for Zen 2 originally back in November, months ahead of the recent Ryzen 3000 series launch, this compiler support was incomplete as it re-used the existing scheduler model and costs table of Znver1. Now though one of SUSE's compiler experts who often works in cooperation with AMD has published the new Znver2 scheduler model and costs table for Zen 2. The updated costs table better reflects the "costs" of moving and loading various registers and different instructions compared to Znver1 so the compiler can make wiser decisions for the most efficient usage. With these updated costs to reflect faster multiplication and 256 vector paths, there is better GNU C Library performance in particular and SUSE developer Jan Hubicka noted that the memory copy performance "wins" even for small blocks.

  • PyCoder’s Weekly: Issue #378 (July 23, 2019)

Graphics: Mesa 19.1.3, RADV and Radeon

  • Mesa 19.1.3
    Mesa 19.1.3 is now available.
    
    In this release we have:
    
    Mostly in fixes for ANV and RADV drivers, as well as NIR backend fixes.
    
    Several of those patches fixe actually crashes with the drivers,
    and a couple of them fix memory leaks.
    
    
    Bas Nieuwenhuizen (3):
          radv: Handle cmask being disallowed by addrlib.
          anv: Add android dependencies on android.
          radv: Only save the descriptor set if we have one.
    
    Caio Marcelo de Oliveira Filho (2):
          anv: Fix pool allocator when first alloc needs to grow
          spirv: Fix stride calculation when lowering Workgroup to offsets
    
    Chia-I Wu (2):
          anv: fix VkExternalBufferProperties for unsupported handles
          anv: fix VkExternalBufferProperties for host allocation
    
    Connor Abbott (1):
          nir: Add a helper to determine if an intrinsic can be reordered
    
    Dave Airlie (1):
          radv: fix crash in shader tracing.
    
    Eric Anholt (1):
          freedreno: Fix assertion failures in context setup in shader-db mode.
    
    Gert Wollny (1):
          softpipe: Remove unused static function
    
    Ian Romanick (4):
          intel/vec4: Reswizzle VF immediates too
          nir: Add unit tests for nir_opt_comparison_pre
          nir: Use nir_src_bit_size instead of alu1->dest.dest.ssa.bit_size
          mesa: Set minimum possible GLSL version
    
    Jason Ekstrand (13):
          nir/instr_set: Expose nir_instrs_equal()
          nir/loop_analyze: Fix phi-of-identical-alu detection
          nir: Add more helpers for working with const values
          nir/loop_analyze: Handle bit sizes correctly in calculate_iterations
          nir/loop_analyze: Bail if we encounter swizzles
          anv: Set Stateless Data Port Access MOCS
          nir/opt_if: Clean up single-src phis in opt_if_loop_terminator
          nir,intel: Add support for lowering 64-bit nir_opt_extract_*
          anv: Account for dynamic stencil write disables in the PMA fix
          nir/regs_to_ssa: Handle regs in phi sources properly
          nir/loop_analyze: Refactor detection of limit vars
          nir: Add some helpers for chasing SSA values properly
          nir/loop_analyze: Properly handle swizzles in loop conditions
    
    Juan A. Suarez Romero (3):
          docs: add sha256 checksums for 19.1.2
          Update version to 19.1.3
          docs: add release notes for 19.1.3
    
    Lepton Wu (1):
          virgl: Set meta data for textures from handle.
    
    Lionel Landwerlin (6):
          vulkan/overlay: fix command buffer stats
          vulkan/overlay: fix crash on freeing NULL command buffer
          anv: fix crash in vkCmdClearAttachments with unused attachment
          vulkan/wsi: update swapchain status on vkQueuePresent
          anv: report timestampComputeAndGraphics true
          anv: fix format mapping for depth/stencil formats
    
    Marek Olšák (1):
          radeonsi: don't set READ_ONLY for const_uploader to fix bindless texture hangs
    
    Samuel Iglesias Gonsálvez (1):
          anv: fix alphaToCoverage when there is no color attachment
    
    Samuel Pitoiset (1):
          radv: fix VGT_GS_MODE if VS uses the primitive ID
    
    Sergii Romantsov (1):
          meta: memory leak of CopyPixels usage
    
    Timothy Arceri (1):
          mesa: save/restore SSO flag when using ARB_get_program_binary
    
    Vinson Lee (1):
          meson: Add dep_thread dependency.
    
    Yevhenii Kolesnikov (1):
          meta: leaking of BO with DrawPixels
    
    git tag: mesa-19.1.3
    
  • Mesa 19.1.3 Led By Fixes For Intel & Radeon Vulkan Drivers

    If you are sticking to stable versions of Mesa, the Mesa 19.1.3 point release is out today as the latest and greatest version of this collection of open-source graphics drivers. Changes for Mesa 19.1.3 are led by the Intel "ANV" and Radeon "RADV" Vulkan drivers. Among those Vulkan driver fixes are taking care of Android dependencies for ANV, external buffer properties fixes for Intel, a crash in shader tracing for RADV, and various other fixes.

  • RADV Lands Binning Support For GFX10/Navi For Faster Vulkan Performance

    The latest change for Mesa 19.2 to better the new Radeon RX 5000 "Navi" series support is binning in the RADV driver. Following all the prep work, Bas Nieuwenhuizen flipped on the binning functionality within the RADV Vulkan API driver for GFX10 (Navi). With testing Talos Principle as a test scenario for primitive binning, Bas found the frame-rates to improve by about 6%. That's not too surprising and most Vulkan games should see frame-rates improve by about a couple of percent if the Vega primitive binning numbers are any indicator from when that support landed in RADV.

  • Radeon Cauldron 1.0 Released As AMD's New SDK Framework

    Radeon Cauldron 1.0 is AMD's new graphics SDK framework for developing Vulkan and Direct3D 12 demos/prototypes/samples. The GPUOpen developers describe this new framework as "like having a simplified game engine that you can learn and modify in little time." Radeon Cauldron makes it easy to load up glTF 2.0 models and to display in either Vulkan or D3D12, flexibility to encompass more graphics features over time, is written using vanilla C++, and has already been used by different teams within AMD.