NVIDIA is working on adding HEVC/H.265 video decoding support to VDPAU.
NVIDIA developers are extending the "Video Decode and Presentation API for Unix" interfaces to allow the HEVC/H.265 requirements. The work aims to enable hardware-accelerated decoding of HEVC content under VDPAU and to provide a reference implementation for this video decoding. José Hiram Soltren, the developer that worked on this support, is also working on a HEVC decode patch for FFmpeg and MPlayer based upon the new API.
For those excited about the recent working Radeon R9 290 "Hawaii" Gallium3D support, a number of bug-fixes were committed in recent hours to Mesa for bettering the support for those wishing to use this open-source AMD Linux driver for their ultra high-end graphics hardware.
Open-source AMD Linux users wishing to use a Hawaii GPU will still need to utilize the patches that will not be queued up until the Linux 3.17 kernel (along with updating their Radeon microcode files) but the RadeonSI Gallium3D Mesa improvements are starting to hit the mainline tree.
Starting out the last week of July's Linux benchmarking on Phoronix is a fresh comparison of several NVIDIA GeForce graphics cards when comparing the performance of the latest open-source Nouveau driver against the latest NVIDIA proprietary Linux graphics driver. While the Kepler cards now support GPU re-clocking, the results aren't quite ideal yet.
AMD Hawaii support works with GLAMOR (both with the external library and the internal support found in X.Org Server 1.16), is running a variety of Steam games, etc. As a word of caution, MSAA might be one of the currently broken Hawaii features unless additionally applying a libdrm patch. Among the titles people are reportedly trying with the Hawaii GPU on RadeonSI Gallium3D include Civilization 5, Half-Life 2, Metro: Last Light, Portal 2, and XCOM: Enemy Unknown. The performance on the open driver is said to be satisfactory in most situations but with XCOM for instance the frame-rate on a R9 290 class GPU is under ten frames per second and there's also issues with GPU stalls. A big problem reported by a user comes down to very poor performance in playback of video streams, such as from Twitch.
For the better part of a year, the X.Org Foundation has been evaluating a possible merger with SPI. That work is still ongoing and could be put up for a vote in the weeks ahead.
At yesterday's X.Org Board of Directors' meeting (IRC log), an update was shared and comes down to Keith Packard working with the SPI on constructing a resolution to bring to the SPI board. However, first the X.Org board will need to review the draft, etc. Assuming the SPI board votes in favor of this resolution, the following step is then to take the revised X.Org Foundation member by-laws to a vote by all X.Org Foundation members.
As the second part of our Linux graphics testing this week after a Radeon R600/RadeonSI performance update with the Linux 3.16 kernel and Mesa 10.3-devel are some comparative numbers that include Intel's Haswell HD Graphics and various NVIDIA GeForce graphics cards on the Nouveau driver.
What we have for this article are the benchmarks of an assortment of AMD Radeon and NVIDIA GeForce GPUs (and the integrated HD Graphics of the Core i7 Devil's Canyon processor used for testing all the hardware) with the latest open-source graphics drivers using Linux 3.16 and Mesa 10.3-devel. Ubuntu 14.04 LTS was running on the system with using the Ubuntu Mainline Kernel PPA for the latest kernel and the Oibaf PPA for the updated graphics drivers.
As the first part of an upcoming series of tests benchmarking the latest open-source and closed-source Linux graphics drivers for AMD Radeon and NVIDIA GeForce hardware, here's some benchmark results for several recent Radeon GPUs when tested on the current Git version of the Linux 3.16 kernel and a recent Mesa 10.3-devel snapshot.
Those wishing to see the Raspberry Pi B+ performance benchmarks with a Debian Linux host, the results are available from 1407220-BY-1407183GL47. To see how the results compare against your own Linux systems, with the Phoronix Test Suite you simply need to run phoronix-test-suite benchmark 1407220-BY-1407183GL47 to conduct a quick, fully-automated, side-by-side performance comparison.
I’m sorry for the late announcement. Here’s a tentative schedule for the 3.5 release.
July 21: Branch for 3.5 release.
- All features should be near completion. Changes to complete existing features will be accepted until the end of testing phase I. But they should *not* be major (i.e., large) changes. If they are, they may be rejected.