Meet Radeon™ GPU Detective

Our powerful new tool for developers to help with post-mortem analysis of GPU crashes.

Latest news from GPUOpen

Using HIP RT ray tracing library to accelerate fluid simulation

AMD HIP Ray Tracing
This blog post shows how it is possible to use HIP RT to accelerate neighbor search in Smoothed Particle Hydrodynamics (SPH) simulations.

Sparse matrix vector multiplication – part 1 – AMD lab notes

AMD Lab Notes
Sparse matrix vector multiplication (SpMV) is a core computational kernel of nearly every implicit sparse linear algebra solver. This is the first post in the series covering SpMV.

Introducing HIP RT v2.1 – batch construction for small geometries, transformation query functions, and more

AMD HIP Ray Tracing
HIP Ray Tracing v2.1 adds support for batch construction, global/dynamic stacks, transformation query functions, plus other new features.

Work graphs API – compute rasterizer learning sample

Work graphs learning sample
Learn more about the power of work graphs API in our detailed blog, taking you step-by-step through an example which implements a scanline rasterizer.

Jacobi Solver with HIP and OpenMP offloading – AMD lab notes

AMD Lab Notes
In this blog, we explore GPU offloading using HIP and OpenMP target directives and discuss their relative merits in terms of implementation efforts and performance.

Radeon™ Raytracing Analyzer 1.3 is now available

Radeon Raytracing Analyzer v1.3 adds the ability to inspect casted rays in 3D and examine ray traversal statistics. Check it out now!

Radeon™ Memory Visualizer 1.7 is out now

Radeon™ Memory Visualizer v1.7 introduces one of the most anticipated new features: improved support for aliased resources in the Resource overview pane. Read on for details of this features and more.

Radeon™ GPU Profiler 1.16 is here!

Radeon GPU Profiler v1.16 updates the GPU ISA disassembly view, adds improved dark mode support, and more.

Creating a PyTorch/TensorFlow Code Environment on AMD GPUs – AMD lab notes

AMD Lab Notes
The machine learning ecosystem is quickly exploding and this article is designed to assist data scientists/ML practitioners get their machine learning environments up and running on AMD GPUs.

Let’s talk about performance…

Building something amazing on DirectX®12 or Vulkan®? How about Unreal Engine?

Obviously you wouldn’t dream of shipping without reading our performance and optimization guides for Radeon, Ryzen, or Unreal Engine first!

Performance

Memory

Shaders

Ray tracing

We aim to provide developer tools that solve your problems.

To achieve this, our tools are built around four key pillars: stability, performance, accuracy, and actionability.

AMD FidelityFX technologies

Our series of optimized, shader-based features aimed at improving rendering quality and performance.  

Our series of optimized, shader-based features aimed at improving rendering quality and performance.  

Now all available as part of the AMD FidelityFX SDK!

FidelityFX features are designed to be easy to integrate in Vulkan® or DirectX®.

We ship HLSL and/or GLSL versions of the effects to allow you to target any API.

AMD Radeon ProRender

Content creators are welcome on GPUOpen! 

Discover the AMD ProRender SDK and plugins for popular content creation tools.

Developing on Unreal Engine?

Our easy to integrate patches help you get more performance and fantastic new features with minimum effort.

Let’s build better…

From managing memory with DirectX®12 or Vulkan®, to asking low-level questions of the driver, our extensive set of SDKs put you in the driving seat.

AMD GPUOpen

Just want to browse?

Have a look through our site directory to see just how much stuff we have!