Category: NVIDIA CUDA / GPU Programming

A brief discussion of Vulkan vs. OpenGL and why I believe you should learn Vulkan over OpenGL going forward. I also talk briefly about OpenGL CUDA interop and the computer graphics industry as a whole. Enjoy! OpenGL CUDA Interop: https://amzn.to/2SKPH61 […]

A very basic overview of how to use NVIDIA Nsight Compute to optimize your CUDA kernel.

A very basic video walkthrough of how to use Nsight Systems to help in optimizing your application. Nsight Systems is software from NVIDIA and is mainly intended to work with the NVIDIA graphics cards and the CUDA programming language. This […]

A new video tutorial on OpenGL CUDA Interoperability (95+ minutes long) is here! This tutorial will be based on a Windows machine and assumes you have CUDA Toolkit 10.1 installed on your machine. The tutorial includes example code and walks […]

An example of creating world-class graphics with the assistance of CUDA & OpenGL. Some serious processing!

Mark Harris from NVIDIA recently gave some advice on making your application faster by using cudaDeviceGetAttribute() instead of udaGetDeviceProperties() in specific cases. Check out the full details here: https://devblogs.nvidia.com/cuda-pro-tip-the-fast-way-to-query-device-properties/

If code in a .cu file calls CUDA Runtime API functions but contains no ‘__device__’ code, rename the file to .cpp and compile it with the host compiler. You’ll get faster compilation time, which adds up in a large project. […]

Learn how to use CUDA Graphs to make your application run faster and more efficiently. This video walkthrough shows you how to create CUDA Graphs by the Stream Capture Method and the Explicit API method. It also includes source code.

DeepStream SDK 4.0 is out now: https://developer.nvidia.com/deepstream-sdk It can run on Tesla T4,  V100, Jetson TX1, TX2, Nano and AGX Xavier. Enjoy!