Now that the Nouveau, Radeon, and LLVMpipe graphics drivers have been tested under Mesa 8.0, what is left? The Intel DRI driver, of course! The open-source Sandy Bridge Linux graphics support is shining with Mesa 8.0 thanks to OpenGL 3.0 support and measurable performance improvements. Intel Ivy Bridge is also ready to run under Linux.
Patches finally arrived last week for 2D color tiling in the Radeon R600 Gallium3D driver. The patches were then re-based this past weekend and benchmarked by Phoronix. Will the 2D color tiling patches, which affect the Linux kernel, Mesa, libdrm, and xf86-video-ati DDX make the more recent Radeon graphics cards more competitive under open-source to the Catalyst driver?
After looking last week at the ATI/AMD Radeon Gallium3D performance under Mesa 8.0 and comparing its performance to Mesa 7.11 and the closed-source AMD Catalyst driver, along with the LLVMpipe driver performance, we're now focusing upon the Nouveau Gallium3D implementation that seeks to provide open-source NVIDIA hardware support. This comparison is pitting Nouveau in Mesa 8.0 against Mesa 7.11 and the official NVIDIA Linux driver.
Now that I've shared eight good features and eight bad traits of Mesa 8.0, which is the open-source graphics hardware library that's now supportive of OpenGL 3.0, it's time to begin looking closer at the performance. In this article are benchmarks of ATI/AMD Radeon graphics cards spanning five generations to show how the Mesa 8.0 performance compares to the previous release (Mesa 7.11) and the proprietary Catalyst driver.
Intel's experimental "Sandy Bridge New Acceleration" (SNA) acceleration architecture is a constant work-in-progress that even in the past two weeks over the holidays has received more than 100 changes. How though is this new 2D acceleration architecture fairing these days rather than the stock UXA configuration? In this article are our first Intel SNA benchmarks of 2012 when enabling this architecture.
Back in December I posted historical Intel Sandy Bridge benchmarks looking at the graphics performance over the course of 2011 that this latest-generation of Intel hardware has been supported under Linux. In this article are some similar Intel OpenGL benchmarks of each quarter going back to the end of 2010, but this time it is for the previous-generation Intel Ironlake hardware.
After yesterday's article about the Grinch that stole the Radeon Gallium3D performance, here's three offending commits since Mesa 7.10 that are causing the open-source Radeon Gallium3D driver to run slower than it should.
There are some significant performance drops right now on Mesa master for the forthcoming 7.12/8.0 release concerning the Gallium3D driver for older ATI Radeon graphics processors. The performance of the R300g driver is now setback compared to earlier Mesa releases.
If after reading about the recent Nouveau DRM improvements, you were hoping the kernel driver updates fixed-up the situation for the Fermi "NVD0" graphics processors, you are sadly mistaken: it is still a busted mess.
After re-testing every AMD Catalyst driver from 2011, the tables have now turned to do the same for the NVIDIA binary graphics drivers from 2011 as the year comes to an end.
With AMD having published the Catalyst 11.12 driver yesterday, the year is now complete as far as their graphics drivers are concerned. As such, for the sixth year, it's time for the year-in-review articles looking at how the NVIDIA and AMD GPU drivers have matured over the past twelve months in terms of features and OpenGL performance.
A thorough performance look at the Intel Core i7 3960X "Sandy Bridge" Extreme Edition processor will be published very soon, but in this article are some benchmarks of using Gallium3D's LLVMpipe driver on this six-core processor with Hyper Threading.
The year began with Intel launching their "Sandy Bridge" processors. While the CPU performance was very impressive for these latest-generation Intel processors, the graphics performance under Linux was a problem. The drivers were not ready in time. Well, they actually were technically available, but in Git source form and not easy for Linux desktop customers. There were also some initial hurdles in the Sandy Bridge Linux graphics support. However, over the past year, the Intel OSTC developers working on the open-source graphics support have dramatically improved the situation. As this article recaps the performance over the past year, Sandy Bridge is now rocking under Linux and Ivy Bridge is ready to go.
What happens if you build Mesa/Gallium3D with LLVM's Clang compiler or the LLVM DragonEgg plug-in with GCC? It has been asked before, so here is an answer.
It's been a while since last looking at the state of power management for Radeon GPUs, but here's an updated look at the various options surrounding power management for modern ATI/AMD graphics processors and their effectiveness. Various drivers, graphics cards, and tuning options are compared.
For those that are thinking about trying out the Sandy Bridge New Acceleration (SNA) architecture option for 2D graphics acceleration by the xf86-video-intel X.Org Linux driver, here are some benchmarks from the recent xf86-video-intel 2.17 release.
Earlier this month I showed the Intel graphics performance hasn't improved much in the Linux 3.2 kernel (but there might be a boost when RC6 is flipped on), but how is this new kernel shaping up for NVIDIA hardware owners wishing to use the open-source and reverse-engineered Nouveau driver? In this article are some benchmarks of the Nouveau DRM driver from recent Linux releases.
As reported earlier this week, HiZ support is now ready for Intel Sandy Bridge graphics under Linux with the Mesa driver. A set of more than three dozen patches were published that finalize this support for Sandy Bridge (Gen6) while the Ivy Bridge (Gen7) support is almost there but there are some performance regressions still being worked out by Intel's OSTC developers.
Back in September I provided the most comprehensive AMD Radeon Linux graphics comparison that took 28 graphics cards from all supported ATI/AMD Radeon product families and tested them under Linux using the latest Catalyst driver as well as the open-source Mesa/Gallium3D driver. In this article is a similar comparison on the NVIDIA side as I take most of the GeForce graphics cards at my disposal and try them under the NVIDIA binary Linux driver and the community-developed open-source "Nouveau" driver. Not only is the OpenGL performance looked at for multiple generations of NVIDIA hardware, but the thermal and power consumption is compared too. In certain OpenGL workloads, the open-source Linux driver is now faster than NVIDIA's own driver for select graphics cards in a fair comparison, but overall the NVIDIA blob still reigns supreme.
While KVM may be very fast for Linux virtualization, one of the areas where VMware and VirtualBox are superior is when it comes to the ability to provide hardware-accelerated 2D/3D support to guest virtual machines that ultimately is passed onto the host and its graphics card / driver. In this benchmark is a look at the gaming performance of Oracle's VM VirtualBox 4.1 when using their "Chromium" driver to enable guest Linux OpenGL acceleration.
A few days back when testing the Linux 3.1 kernel with Intel's Sandy Bridge hardware and then the Intel RC6 power-savings support, I also ran some updated benchmarks of SNA, the new Intel acceleration architecture available from their graphics driver.
In September the 2011 Linux Graphics Survey came to an end, but due to Oktoberfest, AMD Bulldozer Linux testing, OpenBenchmarking.org developments, and other matters, I didn't have time to look at the survey results until this weekend when getting ready for the Ubuntu Developer Summit. Here's the 2011 results looking at what Linux desktop end-users are running when it comes to graphics cards and drivers as well as their key concerns.
The Linux 3.1 kernel was released earlier this week and it further enhances the Intel Sandy Bridge graphics support while also prepping the open-source kernel driver for Intel's next-generation Ivy Bridge processors.
While there's many ongoing improvements for Intel's Sandy Bridge graphics and the next-generation Ivy Bridge graphics within the Linux kernel, Mesa, and xf86-video-intel (namely the SNA acceleration for the DDX), here's some benchmarks from two older Intel systems using the latest Linux 3.1 kernel to see if there are any improvements there.
Here's something interesting or perhaps odd: AMD has been porting the open-source Radeon Linux driver to Windows Embedded Compact 7 (WEC7) as its graphics driver.
If you are not taken by today's release of Ubuntu 11.10 "Oneiric Ocelot", here are some new Radeon graphics benchmarks comparing the Catalyst driver on Ubuntu 11.10 to the latest R600g driver from Mesa 7.12-devel on the soon-to-be-released Linux 3.1 kernel.
Tom Stellard, the former Google Summer of Code student who worked on ATI R300 GLSL compiler improvements and a new register allocator, has been looking into the area of Radeon OpenCL support while now being employed by AMD. However, Tom is working on other open-source Radeon work too. Recently he made improvements to the R300g driver's instruction scheduler to make better use of the texture semaphore.
While there are still several days left of this year's Oktoberfest, to take a short break this morning from benchmarking the wonderful beer, food, and Bavarian females, here are benchmarks of the new Intel HiZ Linux support. Just a few days ago a new, nearly ready patch-set was published for implementing hierarchical Z support within Intel's Mesa DRI driver.
While it's rare for a few days to pass at Phoronix without pulling the latest development code for Mesa / Gallium3D and the Linux kernel DRM in order to run updated Radeon, Intel, and Nouveau Linux graphics benchmarks, LLVMpipe isn't benchmarked as commonly. LLVMpipe is the new CPU-based software acceleration method for the Gallium3D that leverages the Low-Level Virtual Machine to provide better performance than the classic Mesa software rasterizer or Softpipe. Fortunately, upon running a brand new set of tests, the results show a bit more promise but there is still much work ahead.
While AMD's open-source strategy was announced on Phoronix on 7 September 2007, it was on 17 September of the same year that the Novell/SUSE developers did their first public release of their xf86-video-radeonhd driver. This was the X.Org driver created by the Novell Linux engineers in months prior for R500 and R600 GPUs. Here is some special reading -- a letter that was volleyed from Novell to AMD that kicked off this entire process -- to celebrate what would have been the fourth birthday of this open-source Linux driver.
770 display drivers articles published on Phoronix.