A few days back I wrote about being able to finally get the Metro Redux game benchmarks running in an automated manner under Linux to the point that we're now able to test it with the Phoronix Test Suite. With Metro 2033 Redux and Metro Last Light Redux now running well for our testing purposes, I've carried out performance tests of these two games with twenty-two AMD Radeon and NVIDIA GeForce graphics cards on Linux. Besides looking at the normal FPS result there's also frame latency metrics, power consumption data for each of these graphics cards, performance-per-Watt metrics, and GPU thermal results. If you're wondering what graphics card works best for your needs for OpenGL 4.x Linux gaming, here's an interesting look with the Metro Redux titles that premiered on Steam for Linux back in December.
In upgrading to the new ThinkPad X1 Carbon Broadwell ultrabook, I'm debating whether to switch back to Fedora after having used Ubuntu for a number of years on my main production system after some falling out with a few less then stellar Fedora Core releases back in the day (of course, on test systems, there's plenty of Fedora around here but this is just about deciding on my next main OS for business tasks). In waiting for the new Broadwell ultrabook, I've been running some fresh Ubuntu and Fedora Linux tests on some other laptops/ultrabooks in the office.
With the Linux 3.19 kernel stabilizing nicely, here's a first look at the open-source AMD Radeon graphics performance using this new kernel that will be officially released in the weeks ahead. The Linux 3.18 kernel was compared to the latest Git code of Linux 3.19 for several different AMD Radeon HD series and Rx 200 series graphics cards.
This weekend I got around to trying out the GeForce GTX 970 and GTX 980 "Maxwell" graphics cards with the Linux 3.19 kernel now that there's initial support for these new GPUs via the open-source Nouveau DRM driver.
As part of the big DRM pull request for Linux 3.19, most of the Nouveau driver work was "behind the scenes" with not too much to excite end-users besides initial support for the newer Maxwell GPUs. This support in Linux 3.19 though is just limited to kernel mode-setting with no hardware acceleration being in place for this release.
This comparison is similar to the three-way NVIDIA GeForce graphics card comparison from Monday but just testing the Maxwell-based GeForce GTX 970 and GTX 980 graphics cards while running the latest binary drivers on Windows and Linux. As with the other end-of-2014 Windows vs. Linux benchmarks, Windows 7 Pro x64 with all available system updates was used and on the Linux side was Ubuntu 14.10. The latest NVIDIA Linux driver is the 346.22 driver version while the latest Windows version at the time of testing was the 347.09 driver release.
With last week having delivered fresh benchmarks of AMD's Catalyst Linux vs. Windows drivers (and before that an updated Intel Linux vs. Windows OpenGL comparison to end out 2014), here's some updated NVIDIA Linux vs. Windows benchmarks to compare the GeForce graphics drivers at the end of 2014. Three different graphics cards were used in benchmarking the latest NVIDIA Linux vs. Windows performance with the proprietary graphics drivers followed by also having the latest open-source NVIDIA/Nouveau driver results.
It's been a while since I've last tried out the Git code for the next-generation PHP (phpng) that's going to be known as PHP 7.0 when released likely later this year.
The next major release of PHP is to be called PHP7 in order to avoid confusion with the now-defunct PHP6 unicode initiative. PHP 7.0 is likely to be released by the end of 2015 per the PHP7 timeline. If the release candidates begin on time starting in June, we could be looking at the official PHP 7.0 release around October of this year. However, it's largely dependent upon the state of affairs at that point with the quality of the code.
Earlier this week I showed benchmarks of AMD's incredible year for their open-source Linux driver and how the open-source Radeon Gallium3D driver moved closer to performance parity with Catalyst. One of the lingering questions though is how does the Catalyst 14.12 Omega Linux driver from December compare to the latest Catalyst Windows driver? Here's some benchmarks looking at the latest open and closed-source drivers on Linux compared to the latest Catalyst Windows release.
It's been a while since last delivering a Windows vs. Linux Catalyst comparison at Phoronix, but found the time to be right for going along with our year-end recaps and performance reviews. Earlier this week I also posted the Intel Windows vs. Linux OpenGL performance comparison. The same Core i7 4790K Haswell system was used with this AMD Linux vs. Windows benchmarking. As shared in that Intel article, Windows 8 was being very unstable on this particular system so for all of the testing I had to revert to running Windows 7 rather than Windows 8.1.