Following a NVIDIA 450 Linux beta with the CUDA 11.0-rc in early June and the more formal NVIDIA 450.51 Linux beta later in June, NVIDIA has now promoted the 450 Linux driver series to stable with today's release of the 450.57 driver build.

NVIDIA appears to have quietly promoted CUDA 11.0 to its stable channel.

NVIDIA has quietly released Video Codec SDK 10 as the newest version of their proprietary video encode/decode implementation designed for their GPUs.

Our recent benchmarks have shown WSL/WSL2 performance on the latest Windows 10 builds to generally be quite good compared to running bare metal Linux. But past the May 2020 Update and on the latest Insider Preview builds is the initial support for GPU acceleration in conjunction with updated Windows graphics drivers. The initial emphasis is on GPU compute with DirectML and for NVIDIA hardware CUDA support as well. Here are a couple CUDA benchmarks that ran gracefully under WSL2 albeit the performance leaves a lot to be desired.

Earlier this month a NVIDIA 450 Linux beta driver popped out as part of the CUDA 11.0 release candidate. Today though is the first public and generally available NVIDIA 450 series Linux driver beta for all users.

NVIDIA today released a new Vulkan beta driver for Linux systems at version 440.66.17.

After announcing the NVIDIA Ampere architecture at last month's virtual keynote, beginning today the NVIDIA A100 PCI Express accelerator is now shipping in GPU compute servers.

NVIDIA has yet to formally announce the 450 Linux driver series in beta or stable form, but the first pre-release builds in the 450 branch did manage to creep out this past week alongside the CUDA 11.0 release candidate.

While the Tegra X1 SoC (Tegra210) has been available for several years, finally with the upcoming Linux 5.8 kernel is a mainline driver contributed by NVIDIA for the video input support.

NVIDIA has been quite aggressive recently with their new Vulkan beta drivers for Windows and Linux with today marking another such release.

NVIDIA on Friday released Nsight Graphics 2020.3 as the newest version of their proprietary tool for profiling and debugging Direct3D / Vulkan / OpenGL / OpenVR software.

One of the interesting patch series initially published back in 2019 by NVIDIA engineer Nitin Gupta was on proactive memory compaction for the Linux kernel while so far in 2020 it hasn't yet been merged but a fifth revision to the work was published today.

NVIDIA Carmel CPU cores that succeeded Denver 2 and found for a while already within Tegra Xavier hardware now has mainline LLVM/Clang compiler support.

While waiting to see what NVIDIA will be doing on the open-source driver front that has been pushed back, NVIDIA made a surprise open-source announcement today.

NVIDIA has released their latest weekly-ish beta update to their Vulkan Linux driver.

NVIDIA has announced MONAI as their newest open-source initiative.

NVIDIA today issued new beta builds of their Vulkan drivers for Linux and Windows.

NVIDIA today released the 440.82 Linux binary display driver as their newest stable update in the current 440 driver series.

The latest LLVM 11 development code has enabled support for NVIDIA CUDA GPU device offloading from 64-bit ARM.

NVIDIA has released a new feature update to their Nsight Graphics standalone developer tool for debugging and profiling applications/games built atop a variety of 3D APIs.

For those looking to work on Blender 3D modeling from a laptop, having a NVIDIA RTX graphics processor can do wonders with the OptiX back-end for dramatically speeding up render times. Here is a look at how the different back-ends compare when running the HP ZBook 17 G6 mobile workstation with Quadro RTX 5000 graphics.

Continuing on with our Blender 2.82 benchmarking for this open-source 3D modeling software update that debuted last month with numerous improvements, here are some fresh benchmarks of the CUDA and OptiX back-ends for NVIDIA GPU acceleration.

The latest setback from Coronavirus / COVID-19 concerns is NVIDIA's flagship GTC conference no longer happening in San Jose later this month.

NVIDIA has issued another stable Linux driver release in their long-term 440 series.

As an obstacle for upstreaming some particularly older NVIDIA Tegra devices (namely those running Android) is that they have GPT entry at the wrong location or lacking at all for boot support. That missing or botched GPT support is because those older devices make use of a NVIDIA proprietary/closed-source table format. As such, support for this proprietary NVIDIA Tegra Partition Table is being worked on for the Linux kernel to provide better upstream kernel support on these consumer devices.

Big "open-source" achievements aren't too common for NVIDIA or Microsoft much less together, but thanks to their open-source work on the DXC DirectXCompiler it's possible to easily convert HLSL DXR shaders to SPIR-V for Vulkan.

Not scheduled to go live until Monday but up this weekend is the NVIDIA 440.58.01 Linux beta driver that offers a few Vulkan updates.

Codeplay announced last year they were working on an open-source layer for running Intel's oneAPI and Data Parallel C++ on NVIDIA GPUs and as part of that supporting Khronos' SYCL on NVIDIA hardware. Today they revealed more details on this achievement and new software layer.

NVIDIA has kicked off February by releasing the 440.59 Linux driver as their newest stable driver.

NVIDIA has sent out word that they no longer plan to issue anymore driver updates for their 340 series Linux legacy branch.