AMD GPUOpen
@gpuopen.bsky.social
1.4K followers 35 following 47 posts
We're the #gamedev home for AMD FidelityFX™ effects inc. FSR, Radeon™ Developer Tool Suite, VMA, AMF + more.
Posts Media Videos Starter Packs
gpuopen.bsky.social
Train a reinforcement learning agent in #UnrealEngine with the new AMD Schola v1.3 to control a Raspberry Pi-powered 🤖 robot.

➡️ Learn how to deploy AI models from simulation to real-world #robotics with this tutorial: gpuopen.com/learn/sim-to...
Sim-to-real in AMD Schola - AMD GPUOpen
Replicating a physical line-following device in Unreal® Engine and training it with reinforcement learning using AMD Schola.
gpuopen.com
gpuopen.bsky.social
Exciting news for developers working with our AMD DGF geometry compression technology!

🗜️ The DGF SDK has been updated with a new animation-aware encoding pipeline, bug fixes, and improvements.

👉 Learn more and download: gpuopen.com/learn/animat...
Animating geometry with AMD DGF - AMD GPUOpen
The AMD DGF SDK has been updated with improvements and new features, including the addition of an animation-aware encoding pipeline.
gpuopen.com
gpuopen.bsky.social
HIP RT 3 is here, bringing optimizations that use AMD RDNA 4 #raytracing 🧊 features to boost performance on this architecture.

📥 Download now: gpuopen.com/hiprt/?utm_s...
HIP Ray Tracing - AMD GPUOpen
HIP RT is a ray tracing library for HIP, making it easy to write ray tracing applications in HIP.
gpuopen.com
gpuopen.bsky.social
You asked for smarter upscaling, so we taught pixels to think with AMD RDNA 4 architecture 🤔

✨ Introducing AMD #FSR4: our ML-based upscaler for max upscaling quality.

✨ Find it in #FidelityFX SDK 2.0 + our new #UnrealEngine 5 plugin!

➡️ Meet FSR 4: gpuopen.com/learn/amd-fs...
AMD FidelityFX Super Resolution 4 now available on GPUOpen - AMD GPUOpen
Discover AMD FSR 4, our cutting-edge ML-based upscaler (with UE5 plugin) which delivers significant image quality improvements over FSR 3.1.
gpuopen.com
gpuopen.bsky.social
You trained us well. We learned you wanted to see AMD #FSR 4 on GPUOpen 🫣

🟥 Introducing AMD #FidelityFX SDK 2.0: the launchpad for our neural rendering tech, including cutting-edge #FSR4 and upcoming Redstone features!

➡️ Meet the FidelityFX SDK 2: gpuopen.com/learn/amd-fi...
AMD FidelityFX SDK 2.0 launches our neural rendering technologies for developers - AMD GPUOpen
Learn about the launchpad for our ML-based rendering technologies, inc. AMD FSR 4 and upcoming FSR Redstone features.
gpuopen.com
gpuopen.bsky.social
Join us at #SIGGRAPH2025 to attend our exciting courses and presentations, including:

✨GPU Work Graphs on Sunday, Aug 10
✨Demystifying Noise: Role of Randomness in Generative AI on Wednesday, Aug 13

📅 See our schedule: s2025.conference-schedule.org/organization...
Organization - SIGGRAPH 2025 Conference Schedule
s2025.conference-schedule.org
gpuopen.bsky.social
AMD FSR Redstone is our upcoming suite of ML-powered AMD #FidelityFX neural rendering technologies.

🟥 Get your game Redstone ready with #FSR 3.1.4 🟥

We've also updated our #UnrealEngine FSR 3 & Radeon Anti-Lag 2 plugins to support UE 5.6.

Details ➡️ gpuopen.com/learn/amd-fs...
Get ready for FSR Redstone with AMD FSR 3.1.4 and our new UE 5.6 plugin - AMD GPUOpen
Integrate AMD FSR 3.1.4 to get ready for FSR Redstone, updated Unreal Engine 5.6 plugin now available.
gpuopen.com
gpuopen.bsky.social
Working with our block-based geometry compression technology, AMD Dense Geometry Format?

🚀 Improve performance and lower memory costs with the new #Vulkan extension for DGF, supported by our latest preview driver and sample.

👉 Learn more and download: gpuopen.com/learn/dense-...
AMD releases Vulkan support for Dense Geometry Format - AMD GPUOpen
We’ve released a provisional Vulkan® extension for AMD Dense Geometry Format (DGF), our block-based geometry compression technology.
gpuopen.com
gpuopen.bsky.social
Now you can easily use #TressFX 5.0 🦱 hair or fur in Unreal Engine 5 with our latest #UE5 patch.

🆕 This update adds Lumen software/hardware #raytracing support, deep shadow approximation, and more.

📥 Read the blog and download the patch: gpuopen.com/learn/tressf...
AMD TressFX 5.0 for Unreal Engine 5 is now available - AMD GPUOpen
AMD TressFX 5.0 has been updated for high-quality simulation and rendering of realistic hair and fur in Unreal Engine 5.
gpuopen.com
gpuopen.bsky.social
Our #opensource AMD Interactive Streaming SDK has been updated to support 🐧 #Linux.

🎮 #FOSS developers can now build custom low-latency streaming solutions for cloud gaming, VDI, and embedded apps on AMD GPUs.

📥 Download the update: gpuopen.com/news/streami...
AMD Advanced Interactive Streaming SDK - AMD GPUOpen
The AMD Interactive Streaming SDK provides developers with a set of building blocks and samples that allow to easily create custom low-latency streaming solutions for cloud gaming, VDI, and embedded a...
gpuopen.com
gpuopen.bsky.social
Our research on Neural Supersampling and Denoising (NSSD) has been updated.

💥 We've added a lightweight multi-branch and multi-scale feature extraction network for superior visual quality.

Our recently expanded blog has all the details 👉 gpuopen.com/learn/neural...
Neural Supersampling and Denoising for Real-time Path Tracing - AMD GPUOpen
Read our research for a neural denoising and supersampling technique, with the aim of achieving real-time path tracing.
gpuopen.com
gpuopen.bsky.social
Our research in generative AI is paving the way for a new era of 🧠 neural rendering.

✨We're sharing how you can leverage diffusion-based techniques to generate indirect illumination (GI) effects ✨

Find out more in our latest article on GPUOpen:
gpuopen.com/learn/genai-...
Generative AI model for Global Illumination effects - AMD GPUOpen
Our generative AI research leverages diffusion-based techniques to generate indirect illumination effects for the new era of neural rendering.
gpuopen.com
gpuopen.bsky.social
💡 Do you want to learn more about using the matrix cores of AMD RDNA™ 4 architecture GPUs?

Our quick how-to guide shows you how to:

✅ Use WMMA intrinsics on RDNA 4.
✅ Implement a simple MLP.

👉 Read the guide: gpuopen.com/learn/matrix...
Using the Matrix Cores of AMD RDNA 4 architecture GPUs - AMD GPUOpen
This article is a quick how-to guide for using the WMMA intrinsics with our new AMD RDNA™ 4 architecture GPUs.
gpuopen.com
gpuopen.bsky.social
We've just released the latest #FSR 3.1.4 in our updated plugin for #UnrealEngine 5, ready to drop into your game! 🎮

🔥 Easily get your game ready for FSR 4
🔥 New upscaler CVars and multiple bug fixes

📥 Download now! gpuopen.com/learn/ue-fsr...
AMD FidelityFX Super Resolution 3.1.4 Unreal Engine plugin guide
Download the AMD FSR 3.1.1 plugin for Unreal Engine, and learn how to install and use it.
gpuopen.com
gpuopen.bsky.social
🕵🏻 AMD Radeon GPU Detective (RGD), our tool for analysis of GPU crashes ⚠️ adds support for:

🔎 Debug information from the DirectX® Shader Compiler (DXC) ...
🔎 ... that enhances the crash analysis with additional helpful info.

👉 More details: gpuopen.com/learn/enhanc...
(🧵2/2)
gpuopen.bsky.social
🛠️ We’re pleased to announce that the AMD Radeon Developer Tool Suite has been updated with:

🆕 Support for additional @AMDRadeon RX 9000 series GPUs.
🆕 Enhanced features and updates to all the tools.

👉 Learn more: gpuopen.com/learn/radeon...
(🧵1/2)
gpuopen.bsky.social
Join the new AMD Developer Community Discord today to:

💥 Connect with fellow developers and AMD staff
💥 Ask questions and share ideas about #GameDev, #AI dev, and more!

👉 Sign up now: discord.gg/amd-dev
AMD Developer Community Discord
gpuopen.bsky.social
Join AMD SW Engineers Mehdi and Alex tomorrow for their talk at Centre for AI & Society at York University, to learn more about AMD Schola, our library for developing ML-driven agents in #UnrealEngine

▶️ Talk: www.yorku.ca/research/cai...
▶️ Schola: gpuopen.com/amd-schola/?...

#AI #Research #gamedev
Events and Seminars - Welcome to the Centre for AI & Society at York University
Stay up to date with events hosted and sponsored by York University's Centre for AI & Society.
www.yorku.ca
gpuopen.bsky.social
We’ll also be presenting our LSNIF research at the GHOST Day: Applied Machine Learning Conference shortly:

ghostday.pl#agenda
GHOST Day: Applied Machine Learning Conference
Applied Machine Learning Conference
ghostday.pl
gpuopen.bsky.social
End your #gamedev week on a high note 🙌 - download the AMD #FidelityFX SDK 1.1.4 with some new updates:

💥 FSR 3.1.4 with reduced upscaler ghosting in newly disoccluded pixels.

💥 Updates to Brixelizer GI and Breadcrumbs.

📥 Get the update: gpuopen.com/amd-fidelity...
AMD FidelityFX™ SDK
The AMD FidelityFX SDK is our easy-to-integrate solution for developers looking to include FidelityFX features into their games.
gpuopen.com
gpuopen.bsky.social
#Raytracing BVH traversal struggles on GPUs due to irregular control flow.

🎇 Our Locally-Subdivided Neural Intersection Function (LSNIF) is an ML-based alternative for optimized #rendering.

👉 Join us at @i3dsymposium.org for our talk!

Learn more: gpuopen.com/learn/using_...

#I3D2025
Using Neural Networks for Geometric Representation
Explore how Neural Intersection Functions (NIF) and the enhanced LSNIF are poised to reshape ray tracing by replacing traditional BVH traversal with efficient, GPU-friendly neural networks for acceler...
gpuopen.com
gpuopen.bsky.social
After years of continuous development, we just tagged a new release of D3D12 Memory Allocator library v3.0.0 🎉

🏷️ So if you want the latest and greatest D3D12MA, or just more info - head over to GPUOpen!

#Direct3D #DirectX12

gpuopen.com/d3d12-memory...
D3D12 Memory Allocator
The D3D12 Memory Allocator (D3D12MA) is a C++ library that provides a simple and easy-to-integrate API to help you allocate memory for DirectX®12 buffers and textures.
gpuopen.com
gpuopen.bsky.social
🔼 Triangles may have won the hardware-acceleration rendering battle ... 🔼

⏹️ ... but the method in our latest blog enables emulation of quadrilateral rasterization, through bilinear interpolation of quadrilateral vertices! ⏹️

See how it squares up: gpuopen.com/learn/biline...
Bilinear interpolation on a quadrilateral using Barycentric coordinates
A new algebraic method for hardware-accelerated bilinear interpolation on convex quadrilaterals is presented, using the Barycentric coordinate feature of modern GPUs.
gpuopen.com