With now being able to benchmark ArrayFire via the Phoronix Test Suite, I've been having fun running a number of OpenCL graphics card tests with the 300+ available AF tests. The tests over the past week have been using the NVIDIA Linux driver while here are our first Radeon benchmark results using the AMDGPU-PRO driver stack.
Earlier this week I posted some benchmarks of GPUOpen's new Radeon Open Compute ROCm OpenCL stack that premiered last month and they are working to make completely open-source. In those initial benchmarks I compared the ROCm 1.4 OpenCL performance to the existing AMDGPU-PRO OpenCL implementation on Linux. For those wondering how these two Radeon OpenCL stacks compare to NVIDIA, here are some fresh benchmarks.
One would think the graphics of a Core i5 7600K "Kaby Lake" processor would be faster than the Core i5 6600K "Skylake" or even a Core i5 6500, but that's not always the case with the current state of the Linux driver support for the newest-generation Intel hardware.
With running a ton of end of year benchmarks for showing the latest Linux graphics driver performance at the end of 2016, it's mostly focused upon OpenGL and Vulkan, but for those desiring some fresh NVIDIA CUDA numbers, here they are for your viewing pleasure.
Unless you want your graphics card to keep you warm this winter, here's a big comparison of AMD Radeon and NVIDIA GeForce graphics cards under Linux looking at their performance-per-Watt using the latest OpenGL Linux drivers as of the end of 2016. A few days back I posted a 31-way GeForce/Radeon Linux comparison looking at the raw performance with each company's latest Linux drivers going back to the Fermi and R700 days while for this article is looking at the system power consumption and power efficiency for this mass assortment of GPUs.
Last week I published some fresh AMD Linux 4.9 + Mesa 13.1-dev benchmarks on many different AMD Radeon GPUs going all the way back to the Radeon HD 4800 series days. Today those numbers are being complemented by an extensive NVIDIA GeForce Fermi / Kepler / Maxwell / Pascal comparison to make up a 31-way NVIDIA/AMD Linux OpenGL performance comparison. If you are curious how the NVIDIA and AMD Linux performance is with the very latest drivers and going back several hardware generations, this holiday article is for you.
The latest installment of our year-end benchmarks is focusing upon the performance of the NVIDIA Linux driver against the open-source Radeon Vulkan (RADV) driver found within Mesa 13.1-dev. This comparison is particularly interesting given the continuous flow of improvements into Mesa Git, the NVIDIA 375.26 driver release from last week, the big Dota 2 7.00 update debuted earlier this month, and Croteam's Vulkan improvements have rolled into TTP stable.
This past weekend I published a number of year-end 2016 AMD Linux benchmarks on a wide-range of AMD GPUs going back many generations while using the Linux 4.9 kernel on Ubuntu along with the Mesa 13.1-development code for having the newest open-source Gallium3D drivers. Those results were very interesting and go check them out now if you haven't done so already. For this article is a sub-set of those tests carried out again while monitoring the AC power consumption, GPU temperature, and CPU utilization while also automatically calculating the performance-per-Watt.
With 2016 soon drawing to an end, it's time for all of my year-end recaps now of Linux drivers that I have been doing for the past 12 years. Today are benchmarks of a wide assortment of AMD graphics cards on both R600g and RadeonSI Gallium3D drivers when using Mesa 13.1-dev + LLVM 4.0 SVN and the Linux 4.9 kernel for providing a bleeding-edge look at the open-source AMD Linux graphics performance across hardware going from the Radeon HD 4890 series all the way up through the RX 480 and R9 Fury hardware. Here's a fun look at the OpenGL driver performance across this range of GPUs.
With having published some Darktable OpenCL benchmarks at the beginning of the week with 20 different GPUs (plus more follow-up benchmarks), it didn't take long before requests came in to see some fresh OpenCL Blender benchmarks.
Not that you would normally buy a cheap NVIDIA GeForce graphics card for deep learning tasks like the recently launched GTX 1050 series, as part of running some other fresh CUDA+OpenCL benchmarks I realized I hadn't run any Caffe benchmarks in a while so here are some fresh numbers today. With thirteen NVIDIA GeForce graphics cards including all the consumer GeForce GTX 1000 Pascal cards to date, here are some Caffe benchmarks using the latest NVIDIA 375.10 Linux driver on Ubuntu along with CUDA 8.0 and cuDNN.
Now having had the NVIDIA GeForce GTX 1050 and GTX 1050 Ti graphics cards in my possession for a few days I have some more interesting data to share today compared to just last week's raw OpenGL/OpenCL/Vulkan raw Linux performance data. In this article is a look at the power use and performance-per-Watt of the GeForce GTX 650, GTX 750, GTX 750 Ti, GTX 950, GTX 1050, and GTX 1050 Ti compared to the AMD Radeon RX 460 and RX 470. Additionally, for the newer cards still relevant, there is also performance-per-dollar metrics too.
Yesterday I published the first GeForce GTX 1050 Linux benchmarks with OpenGL, OpenCL, and Vulkan coverage. With now having my hands on the EVGA GeForce GTX 1050 Ti SC GAMING graphics card, here are the first Linux benchmarks of the GTX 1050 Ti graphics card that can be fetched for less than $150 USD.
Earlier this week NVIDIA began shipping the GeForce GTX 1050 graphics cards and our first review is of a Zotac GeForce GTX 1050 Mini. A GeForce GTX 1050 Ti Linux review is still coming up plus some other articles looking at performance-per-Watt and other interesting areas for these low-cost Pascal-based GPUs. Here are results of the latest NVIDIA Linux performance compared to the latest open-source AMD Linux driver with various Radeon GPUs.
In prepping for the GeForce GTX 1050 Linux graphics card reviews this week, I've been re-testing my various AMD and NVIDIA graphics cards atop the very latest driver stacks. As a precursor while waiting for the GeForce GTX 1050 Linux review in the days ahead, here are those fresh benchmarks of the other graphics cards.
With Mesa Git and Linux 4.9 I have done AMDGPU vs. Radeon, AMD OpenGL vs. Vulkan benchmarks, and other tests while in this article is our first side-by-side comparison of AMD Radeon GPUs on their open-source driver against NVIDIA GPUs on Nouveau when using Linux 4.9 Git and Mesa Git as of this week.
After running many OpenGL and Vulkan NVIDIA vs. AMD Linux benchmarks earlier this week, here is a 16-way graphics card comparison when testing the AMD Radeon "Polaris" and NVIDIA GeForce "Pascal" GPUs, among others, on Ubuntu 16.04 LTS and looking squarely at the OpenCL compute performance. Many OpenCL tests plus performance-per-Watt metrics too when using the latest NVIDIA proprietary Linux driver and AMDGPU-PRO.
Yesterday I published early open-source benchmarks of the Radeon RX 470 while today is a full 18-way graphics card comparison including the newly-launched Radeon RX 460 and Radeon RX 470 graphics cards alongside the RX 480 Polaris graphics card. All of the AMD graphics cards tested for this article were running the very latest open-source driver stack on the Linux 4.8 kernel and Mesa 12.1-dev Git.
With my Radeon RX 470 retail unit finally having arrived yesterday, I've been running many benchmarks of this graphics card compared to other AMD and NVIDIA graphics cards under Linux. For your viewing pleasure today is the very tip of the iceberg of many RX 460 and RX 470 Linux benchmarks to be published on Phoronix over the days to come.
To complement yesterday's launch-day GeForce GTX 1060 Linux review, here are some more benchmark results with the various NVIDIA x60 graphics cards I have available for testing going back to the GeForce GTX 460 Fermi. If you are curious about the raw OpenGL/OpenCL/CUDA performance and performance-per-Watt for these mid-range x60 graphics cards from Fermi, Kepler, Maxwell, and Pascal, here are these benchmarks from Ubuntu 16.04 Linux.
Today's the day that we can finally publish NVIDIA GeForce GTX 1060 benchmarks! Today the GTX 1060 begins shipping as NVIDIA's $249 Pascal graphics card to take the Radeon RX 480 head-on. Here's all of the Linux benchmarks you've been waiting to see for the GTX 1060 under Vulkan, OpenGL, OpenCL, and CUDA. compared to various other AMD Radeon and NVIDIA GeForce GPUs on Ubuntu Linux.
One week after the launch of the Radeon RX 480, NVIDIA is lifting the lid this morning on the GeForce GTX 1060 Pascal graphics card with pricing at $249+ USD while delivering GeForce GTX 980 class performance. I already have been testing the GeForce GTX 1060 under Ubuntu Linux, but unfortunately that embargo doesn't expire today... But here's the run-down on all of the technical details on the GTX 1060.
With the Radeon RX 480 Linux review now being out of the way and our various other RX 480 Linux benchmarks, the latest results I have to share with being a benchmarking fanatic are RX 480 results with high-end AMD GPU tests of each generation going back to the Radeon HD 4850/4870 (RV770) days. This article has high-end GPUs from the RX 480 to RX 200, HD 7900, HD 6900, HD 6800, HD 5800, and HD 4800 series compared side-by-side with the latest open-source Radeon Linux graphics driver code. Not only is the raw performance being looked at but the system power consumption was also being polled in real-time for looking at the performance-per-Watt too. For any other benchmarking fanatics curious about the Radeon GPU evolution over the past eight years (RV770 launch in 2008), here are the numbers to enjoy.
After weeks of anticipation, AMD's high-end Radeon RX 480 "Polaris" graphics card is officially launching today! This graphics card starts at just $199 USD (or $239 USD for the 8GB version) and has day-one Linux support! There's available open-source driver support as well as an AMDGPU-PRO update that's expected today for those wanting to make use of this newer hybrid Linux driver stack. I've been testing the Radeon RX 480 under Linux the past week under both driver stacks and have my initial results to share this morning.
If you were amazed by the GeForce GTX 1080 performance under Linux but its ~$699 USD price-tag is too much to handle, the GeForce GTX 1070 is now shipping for $399~449 USD. NVIDIA sent over a GeForce GTX 1070 and I've been putting it through its paces under Linux with a variety of OpenGL, OpenCL, and Vulkan benchmarks along with CUDA and deep learning benchmarks. Here's the first look at the GeForce GTX 1070 performance under Ubuntu Linux.
Last week I published the first Linux review of the GeForce GTX 1080 followed by some performance-per-Watt and OpenGL results from the GTX 1080 going as far back as the 9800GTX, among other interesting follow-up tests with OpenGL/Vulkan/OpenCL. Since then one of the most popular requests has been for doing some deep learning benchmarks on the GTX 1080 along with some CUDA benchmarks, for those not relying upon OpenCL for open GPGPU computing. Here are some raw performance numbers as well as performance-per-Watt in the CUDA space.
Last week when posting my initial NVIDIA GeForce GTX 1080 Linux review the Radeon Linux performance numbers I included were from the latest open-source driver stack, since that's what most Phoronix readers seem interested in as of late given the rapid progress recently of OpenGL 4.x support inside Mesa, the hybrid driver stack also using the AMDGPU kernel driver, etc. But some people expressed curiosity over the AMDGPU-PRO performance relative to NVIDIA particularly with their new GTX 1080 graphics processor. So here is a fresh NVIDIA vs. AMDGPU-PRO graphics card comparison on Linux.
Now that my initial GeForce GTX 1080 Linux review is out the door, I spent this weekend working on a "fun" comparison out of curiosity to see how the raw OpenGL and OpenCL performance has improved over the generations going back to the once-powerful GeForce 9800GTX plus including the top-end cards of the GeForce 600/700/900 Kepler and Maxwell series too.
$699 USD is a lot to spend on a graphics card, but damn she is a beauty. Last month NVIDIA launched the GeForce GTX 1080 as the current top-end Pascal card and looked great under Windows while now finally having my hands on the card the past few days I've been putting it through its paces under Ubuntu Linux with the major open APIs of OpenGL, OpenCL, Vulkan, and VDPAU. Not only is the raw performance of the GeForce GTX 1080 on Linux fantastic, but the performance-per-Watt improvements made my jaw drop more than a few times. Here are my initial Linux results of the Gigabyte GeForce GTX 1080 Founder's Edition.
In part due to the Phoronix 12th birthday this week with running various historical performance comparisons and other interesting benchamrks and in part due to prepping for some long-term comparison data to the Radeon RX 480 launch later this month, for your viewing pleasure this morning are benchmarks testing a variety of graphis cards going back to the Radeon HD 3000 (RV600) series up through the Radeon R9 Fury (Fiji) graphics cards. Enjoy this fun article focusing primarily on the OpenGL performance under Linux over the several generations of ATI/AMD GPUs along with calculating the performance-per-Watt.
236 graphics cards articles published on Phoronix.