Video quality metrics built for filmed videos miss upscaling artifacts in real-time rendering. Video Rater is open-source research for rating game engine output.
🎮 Rate renders vs. ground truth.
🧪 Helps build better metrics.
Get the paper ⬇️
gpuopen.com/learn/publicat…
🎮 Rate renders vs. ground truth.
🧪 Helps build better metrics.
Get the paper ⬇️
gpuopen.com/learn/publicat…
3
7
40
How does @11bitstudios balance creative ambition with technical performance?
🎮 AMD FSR delivers performance & better visuals across a wide range of hardware.
🤝 Collaboration with AMD to help solve development challenges.
Watch the story ⬇️
youtube.com/watch?v=k72QHi…
🎮 AMD FSR delivers performance & better visuals across a wide range of hardware.
🤝 Collaboration with AMD to help solve development challenges.
Watch the story ⬇️
youtube.com/watch?v=k72QHi…
1
3
35
This paper was presented at @EGSympRendering (Eurographics Symposium on Rendering 2026)!
Watch the presentation ⬇️
youtube.com/watch?v=fo6tcb…
#GPU #workgraphs #rendering
Watch the presentation ⬇️
youtube.com/watch?v=fo6tcb…
#GPU #workgraphs #rendering
0
4
19
Missed #UnrealFest26?
Catch up on our time at the event, including:
🎙️ Databiomes' AI voice moderation demo, optimized in collaboration with AMD.
🎮 Epic Games on the State of Unreal keynote.
👥 How the AMD team supports game devs.
Watch the recap ⬇️
youtube.com/watch?v=09H_jE…
Catch up on our time at the event, including:
🎙️ Databiomes' AI voice moderation demo, optimized in collaboration with AMD.
🎮 Epic Games on the State of Unreal keynote.
👥 How the AMD team supports game devs.
Watch the recap ⬇️
youtube.com/watch?v=09H_jE…
0
3
18
Proud to sponsor #UnrealFest26!
Join AMD & Databiomes 2PM CT today:
🧠 "Scaling AI the Smart Way: Memory Footprint and Runtime Efficiency"
⚡ Optimizing AI models for in-game inference.
🎮 Scaling across CPU, GPU, & NPU in Unreal Engine.
More info ⬇️ unrealfest.unrealengine.com/event/46281576…
Join AMD & Databiomes 2PM CT today:
🧠 "Scaling AI the Smart Way: Memory Footprint and Runtime Efficiency"
⚡ Optimizing AI models for in-game inference.
🎮 Scaling across CPU, GPU, & NPU in Unreal Engine.
More info ⬇️ unrealfest.unrealengine.com/event/46281576…
0
2
17
Congratulations to our AMD Developer Community on turning 1! 🎉
🙏 Huge thanks to everyone on Discord & our Community Forums. You’ve helped shape what comes next from AMD for developers.
🔁 See what we’ve built together in our Year 1 recap!
Join now 👉 gpuopen.com/devcommunity/?…
🙏 Huge thanks to everyone on Discord & our Community Forums. You’ve helped shape what comes next from AMD for developers.
🔁 See what we’ve built together in our Year 1 recap!
Join now 👉 gpuopen.com/devcommunity/?…
1
1
18
AMD is proud to support @CDPROJEKTRED's Dziewczyny w Grze! (Girls in the Game!) 💜
Now in its 5th edition, this scholarship & mentoring program helps young women take their first steps into #gamedev 🎮
Celebrating 5 years of dreams and levels unlocked.
youtube.com/watch?v=-fFSfB…
Now in its 5th edition, this scholarship & mentoring program helps young women take their first steps into #gamedev 🎮
Celebrating 5 years of dreams and levels unlocked.
youtube.com/watch?v=-fFSfB…
0
1
19
We're at @XPGamingInc in Toronto! 🍁 #XP26
Come find us at the AMD booth:
👥 AMD Developer Programs & GPUOpen.
🎮 FSR and Schola demos.
🎤 Catch our presentation Friday afternoon.
#gamedev session info ⬇️
xpgamesummit.com/sessions/advan…
Come find us at the AMD booth:
👥 AMD Developer Programs & GPUOpen.
🎮 FSR and Schola demos.
🎤 Catch our presentation Friday afternoon.
#gamedev session info ⬇️
xpgamesummit.com/sessions/advan…
0
4
36
Want to build your own ray tracer or path tracer on AMD GPUs? Our #Raytracing using #HIP paper covers:
🎇 The essential math
🎇 Camera setup, intersections, NEE, BVH + more
🎇 How to accelerate everything with HIPRT
Find it with our other papers: gpuopen.com/learn/publicat…
Direct //t.co/8NMlJ5t3kS
🎇 The essential math
🎇 Camera setup, intersections, NEE, BVH + more
🎇 How to accelerate everything with HIPRT
Find it with our other papers: gpuopen.com/learn/publicat…
Direct //t.co/8NMlJ5t3kS
2
37
195
Join us next week at the Montreal International Game Summit on Nov 11-12.
🎓 Attend our “Advancing AI Game Development with AMD” developer session.
🛠️ Visit our booth to see some of our latest tech demos.
👉 Register now: migs.biz/en/sessions/ad…
#MIGS25 #gamedev
🎓 Attend our “Advancing AI Game Development with AMD” developer session.
🛠️ Visit our booth to see some of our latest tech demos.
👉 Register now: migs.biz/en/sessions/ad…
#MIGS25 #gamedev
2
5
25
Get up to speed with our latest advanced graphics research:
📝 GATE (Geometry-Aware Trained Encoding) for neural rendering.
📝 Real-time procedural GPU tree generation.
📝 Neural Visibility Cache for real-time light sampling.
📥 Download the papers: gpuopen.com/learn/publicat…
📝 GATE (Geometry-Aware Trained Encoding) for neural rendering.
📝 Real-time procedural GPU tree generation.
📝 Neural Visibility Cache for real-time light sampling.
📥 Download the papers: gpuopen.com/learn/publicat…
1
33
132
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_s…
💥 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_s…
12
60
272
🕵🏻 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/enhancin…
(🧵2/2)
🔎 Debug information from the DirectX® Shader Compiler (DXC) ...
🔎 ... that enhances the crash analysis with additional helpful info.
👉 More details: gpuopen.com/learn/enhancin…
(🧵2/2)
0
1
11
🛠️ 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-d…
(🧵1/2)
🆕 Support for additional @AMDRadeon RX 9000 series GPUs.
🆕 Enhanced features and updates to all the tools.
👉 Learn more: gpuopen.com/learn/radeon-d…
(🧵1/2)
3
13
81
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
💥 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
2
15
36
Join us & the Microsoft @DirectX12 team at #GDC2025 for exciting sessions that push the boundaries of #gamedev 🛠️
📈 2.40pm today: our in-depth look at how Work Graphs are redefining GPU programmability!
👉 Learn more about all the DirectX sessions: devblogs.microsoft.com/directx/direct…
📈 2.40pm today: our in-depth look at how Work Graphs are redefining GPU programmability!
👉 Learn more about all the DirectX sessions: devblogs.microsoft.com/directx/direct…
1
19
55
According to the #LunarG 2024 Ecosystem Survey Report, over half of all developers surveyed use our Vulkan Memory Allocator (VMA) library for their #Vulkan development 🎉
Thank you for your trust and support! 🥰
VMA: gpuopen.com/vulkan-memory-…
Survey: lunarg.com/2024-ecosystem…
Thank you for your trust and support! 🥰
VMA: gpuopen.com/vulkan-memory-…
Survey: lunarg.com/2024-ecosystem…
0
10
35
AMD has been providing production support for the TV anime: “Blue Box” / #アオのハコ
AMD Radeon ProRender with #Houdini is being used for the CGI.
🔗 Houdini RPR plugin: github.com/GPUOpen-Librar…
🔗 Learn more about Radeon #ProRender on GPUOpen: gpuopen.com/radeon-pro-ren…
AMD Radeon ProRender with #Houdini is being used for the CGI.
🔗 Houdini RPR plugin: github.com/GPUOpen-Librar…
🔗 Learn more about Radeon #ProRender on GPUOpen: gpuopen.com/radeon-pro-ren…
0
35
82
We've got another exciting technical report to share with you today:
🎓 A Numerically Stable Implementation of the von Mises–Fisher Distribution on S^2
Read the full report via GPUOpen (direct link - 1MB PDF): gpuopen.com/download/publi…
🎓 A Numerically Stable Implementation of the von Mises–Fisher Distribution on S^2
Read the full report via GPUOpen (direct link - 1MB PDF): gpuopen.com/download/publi…
![An image showing plots of sample directions [cos𝜙 sin𝜃,sin𝜙 sin𝜃] in the local frame for vMF distributions. For lowfrequency distribution (upper row) and high-frequency distribution (lower row), Jakob [2012]’s method with single precision (b) generates highly correlated samples due to numerical errors, while ours (c) does not.
First column: (a) Jakob 2012 (double precision) Second column: (b) Jakob 2012 (single precision) Third column: (c) Ours (single precision)](/pic/orig/media%2FGhV6a8FXEAAedNA.jpg)
0
26
88
If you're at #CES2025, head over to the AMD booth!
🔥 We're showing our GI-2.0 AI-assisted real-time path tracing, accelerated by our NEW neural super sampling denoising technique 🔥
If you missed our blog last year, here it is: gpuopen.com/learn/neural_s…
🔥 We're showing our GI-2.0 AI-assisted real-time path tracing, accelerated by our NEW neural super sampling denoising technique 🔥
If you missed our blog last year, here it is: gpuopen.com/learn/neural_s…
🎇 We're excited to share more of our latest #raytracing #research in our latest blog! 🎇
➡️ Read about our new neural denoising and supersampling technique, where we aim to achieve real-time #pathtracing 🔥
Find out how on GPUOpen ⤵️
gpuopen.com/learn/neural_s…
9
33
147
At #Siggraph Asia 2024, we present an unbiased many-light sampling method based on spherical Gaussians.
To estimate the importance of lights, we introduce new high-quality spherical Gaussian convolutions for diffuse and glossy surfaces.
Top table entry:
gpuopen.com/advanced-rende…
To estimate the importance of lights, we introduce new high-quality spherical Gaussian convolutions for diffuse and glossy surfaces.
Top table entry:
gpuopen.com/advanced-rende…
2
44
116
🎇 DMMs (displaced micro-meshes) deliver excellent geometry compression but don't represent animated geometry well 🎇
Our paper explores producing high-quality animated DMMs, showing how to preserve their low memory footprint, BVH build, and update costs:gpuopen.com/download/publi…
Our paper explores producing high-quality animated DMMs, showing how to preserve their low memory footprint, BVH build, and update costs:gpuopen.com/download/publi…
0
41
146
You can watch our #HIP #Raytracing presentation at @HPG_Conf: youtube.com/live/lx1v-Bcza… 📽️
🎇 We've also added some more tutorials for HIP RT, including how to use HIP RT to compute direct illumination using Monte Carlo integration:
github.com/GPUOpen-Librar…
🎇 We've also added some more tutorials for HIP RT, including how to use HIP RT to compute direct illumination using Monte Carlo integration:
github.com/GPUOpen-Librar…
🧵(1/4) We'll be at @HPG_Conf, Denver, July 27! We're very excited to share the four papers we'll be presenting in this thread 🔥
📜 Paper 1 - HIPRT: Our #raytracing framework in HIP
Take a look (28MB): gpuopen.com/download/publi…
Learn more about #HIP RT: gpuopen.com/hiprt/?utm_sou…
0
30
100
@bitsaucetweets Hello! It sounds like you may have already found the environment variables seen below at gpuopen.com/gdc-presentati…
There's a dedicated Steam Deck video here: youtube.com/watch?v=If7Dot… + many more tools-related videos here: gpuopen.com/learn/getting-…
DM if you need anything more! 🙂
There's a dedicated Steam Deck video here: youtube.com/watch?v=If7Dot… + many more tools-related videos here: gpuopen.com/learn/getting-…
DM if you need anything more! 🙂
1
8
36
If you're looking to learn #deeplearning in a #graphics dev context, @boksajak has written an excellent crash course which we're excited to share on GPUOpen 🎉
You'll learn about #neuralnetworks, practical graphics applications, + links to sample code!
gpuopen.com/learn/deep_lea…
You'll learn about #neuralnetworks, practical graphics applications, + links to sample code!
gpuopen.com/learn/deep_lea…
1
26
114
🧵(4/4) @HPG_Conf
📜 Paper 4 - Real-Time Procedural Generation with GPU Work Graphs
This paper shows some of the procedural effects from our #GDC2024 #WorkGraphs demo! 🏰
gpuopen.com/download/publi… (32MB)
Demo: gpuopen.com/learn/gdc-2024…
📜 Paper 4 - Real-Time Procedural Generation with GPU Work Graphs
This paper shows some of the procedural effects from our #GDC2024 #WorkGraphs demo! 🏰
gpuopen.com/download/publi… (32MB)
Demo: gpuopen.com/learn/gdc-2024…
0
20
59
🧵(3/4)
📜 Paper 3 - DGF: A Dense, Hardware-Friendly Geometry Format for Lossily Compressing Meshlets with Arbitrary Topologies.
DGF will do for geometry what block compression did for textures, enabling #raytracing of massive micropolygon geometry 🗜️
gpuopen.com/download/publi…
📜 Paper 3 - DGF: A Dense, Hardware-Friendly Geometry Format for Lossily Compressing Meshlets with Arbitrary Topologies.
DGF will do for geometry what block compression did for textures, enabling #raytracing of massive micropolygon geometry 🗜️
gpuopen.com/download/publi…
5
29
115
🧵(2/4) @HPG_Conf
📜 Paper 2 - H-PLOC: Super-fast, high-quality BVH construction.
Allows for constructing a BVH over millions of triangles in just a few ms. Very useful for efficiently #raytracing a large amount of fully dynamic geometry per frame 🎇
gpuopen.com/download/publi…
📜 Paper 2 - H-PLOC: Super-fast, high-quality BVH construction.
Allows for constructing a BVH over millions of triangles in just a few ms. Very useful for efficiently #raytracing a large amount of fully dynamic geometry per frame 🎇
gpuopen.com/download/publi…
4
12
60
We'll present "Neural Texture Block Compression" @ #EGSR2024 in London.
Nobody likes downloading huge game packages. Our method compresses the texture using a neural network, reducing data size.
Unchanged runtime execution allows easy game integration. egsr2024.uk/programme/
Nobody likes downloading huge game packages. Our method compresses the texture using a neural network, reducing data size.
Unchanged runtime execution allows easy game integration. egsr2024.uk/programme/
10
66
325
✨ Our Microsoft @DirectX12 Work Graphs session at the Advanced Graphics Summit at #GDC2024 created a lot of excitement ✨
Now you can discover how Work Graphs will transform GPU programming for yourself, with the just-released video and slide deck: youtube.com/watch?v=QQP6-J…
Now you can discover how Work Graphs will transform GPU programming for yourself, with the just-released video and slide deck: youtube.com/watch?v=QQP6-J…
4
49
122
We're excited to share our new #research paper: "Efficient Visibility Reuse for Real-time ReSTIR"
◼️ Our technique reuses info from nearby pixels to estimate the visibility, improving shadow edge quality ◼️
Find all links on ARR Group's page on GPUOpen: gpuopen.com/advanced-rende…
◼️ Our technique reuses info from nearby pixels to estimate the visibility, improving shadow edge quality ◼️
Find all links on ARR Group's page on GPUOpen: gpuopen.com/advanced-rende…
3
91
358
✨ And don't miss @domipheus announcing AMD #FSR 3.1 in the video of our "High Performance Rendering in Snowdrop using AMD #FidelityFX Super Resolution 3" session ✨
youtu.be/ojhkxXLRcTU?si…
youtu.be/ojhkxXLRcTU?si…
1
9
53
🧵 Close the case on #GDC24 Wednesday by investigating our #Radeon GPU Detective session 🕵️
@Reg__ & Amit + guest Marco Bouterse (@NixxesSoftware) teach capturing and analyzing AMD GPU crash dumps with RGD.
🗓️ Wed Mar 20 @ 5pm
📌 3001, West Hall
schedule.gdconf.com/session/post-m… (8/8)
@Reg__ & Amit + guest Marco Bouterse (@NixxesSoftware) teach capturing and analyzing AMD GPU crash dumps with RGD.
🗓️ Wed Mar 20 @ 5pm
📌 3001, West Hall
schedule.gdconf.com/session/post-m… (8/8)
0
6
19
🧵 It’s #GDC24 Tools Time with Amit & Chris! Get the latest @AMDRadeon Dev Tool Suite news 🛠️
Plus hear about shader instrumentation with brand-new GPU Reshape, with creator @miguel_oenp (@sd_studios)
🗓️ Wed Mar 20 @ 3.30pm
📌 3001, West Hall
schedule.gdconf.com/session/game-o… (7/8)
Plus hear about shader instrumentation with brand-new GPU Reshape, with creator @miguel_oenp (@sd_studios)
🗓️ Wed Mar 20 @ 3.30pm
📌 3001, West Hall
schedule.gdconf.com/session/game-o… (7/8)
1
4
17
🧵 Bring some sunshine to your #GDC24 afternoon with @diharaw94, as he shines a light on upcoming AMD FidelityFX Brixelizer & Brixelizer GI 🌞
He’ll also update you on the latest AMD #FidelityFX SDK news!
🗓️ Wed Mar 20 @ 2pm
📌 3001, West Hall
schedule.gdconf.com/session/global… (6/8)
He’ll also update you on the latest AMD #FidelityFX SDK news!
🗓️ Wed Mar 20 @ 2pm
📌 3001, West Hall
schedule.gdconf.com/session/global… (6/8)
2
8
47
🧵 If #CPU bottlenecks ruin your day, fix it with @KenMitchellKen’s AMD #Ryzen optimization #GDC24 session 💻
Learn about micro-architecture, optimizations, frequent issues, benchmarking, + more!
🗓️ Wed Mar 20 @ 12.30pm
📌 3001, West Hall
schedule.gdconf.com/session/amd-ry… (5/8)
Learn about micro-architecture, optimizations, frequent issues, benchmarking, + more!
🗓️ Wed Mar 20 @ 12.30pm
📌 3001, West Hall
schedule.gdconf.com/session/amd-ry… (5/8)
1
2
13
🧵 Upscale your #GDC24 experience with @domipheus updating you on AMD #FSR 3 ✨
Guest Hampus Siversson (@UbiMassive) will share FSR 3➡️Snowdrop Engine integration details + how this aided FSR 3 dev work!
🗓️ Wed Mar 20 @ 10.30am
📌 3001, West Hall
schedule.gdconf.com/session/high-p… (4/8)
Guest Hampus Siversson (@UbiMassive) will share FSR 3➡️Snowdrop Engine integration details + how this aided FSR 3 dev work!
🗓️ Wed Mar 20 @ 10.30am
📌 3001, West Hall
schedule.gdconf.com/session/high-p… (4/8)
1
7
23
🧵 Join Lou 👩🔧 & Max 👨🔧 for a #GDC24 deep-dive into mesh shaders + explore new ways of rendering & compressing assets 🗜️
You’ll also learn how to best optimize #meshshaders for #AMD #RDNA 3 architecture!
🗓️ Wed Mar 20 @ 9am
📌 3001, West Hall
schedule.gdconf.com/session/mesh-s… (3/8)
You’ll also learn how to best optimize #meshshaders for #AMD #RDNA 3 architecture!
🗓️ Wed Mar 20 @ 9am
📌 3001, West Hall
schedule.gdconf.com/session/mesh-s… (3/8)
1
5
32
🧵 Don't miss @NIV_Anteru (@AMD) + Shawn Hargreaves (@DirectX12) @ #GDC24 Advanced #Graphics Summit 📈
Discover the impact the new Microsoft Work Graphs API will have on the #GPU programming landscape!
🗓️ Mon Mar 18 @ 1.20pm
📌 303, South Hall
schedule.gdconf.com/session/advanc… (2/8)
Discover the impact the new Microsoft Work Graphs API will have on the #GPU programming landscape!
🗓️ Mon Mar 18 @ 1.20pm
📌 303, South Hall
schedule.gdconf.com/session/advanc… (2/8)
1
4
29
🧵 (3/3) The other paper to be presented:
🎇 2. "Bounded VNDF Sampling for Smith–GGX Reflections"
▶️ asia.siggraph.org/2023/presentat… Wed 13 Dec @ 4:50pm C4.8, Level 4
If you're interested in joining ARR group, talk to the team at #SIGGRAPHAsia2023 or contact details on their papers.
🎇 2. "Bounded VNDF Sampling for Smith–GGX Reflections"
▶️ asia.siggraph.org/2023/presentat… Wed 13 Dec @ 4:50pm C4.8, Level 4
If you're interested in joining ARR group, talk to the team at #SIGGRAPHAsia2023 or contact details on their papers.
0
3
18
🧵 (2/3) Also at #SIGGRAPHAsia2023, ARR group will be presenting two papers.
🎆 1. "Real-time Rendering of Glossy Reflections using #RayTracing and Two-level Radiance Caching"
▶️ asia.siggraph.org/2023/presentat… Tues 12 Dec @ 5.41pm, C4.11, Level 4
Both papers: gpuopen.com/advanced-rende…
🎆 1. "Real-time Rendering of Glossy Reflections using #RayTracing and Two-level Radiance Caching"
▶️ asia.siggraph.org/2023/presentat… Tues 12 Dec @ 5.41pm, C4.11, Level 4
Both papers: gpuopen.com/advanced-rende…
1
3
27
🧵 (1/3) Our Advanced #Rendering Research (ARR) group will be at #SIGGRAPHAsia2023 next week, presenting sessions.
🎓 A course on #GPU Programming Primitives for Computer Graphics
▶️ asia.siggraph.org/2023/presentat… Thurs 14 Dec @ 9am C4.6, Level 4
▶️ More info: gpu-primitives-course.github.io
🎓 A course on #GPU Programming Primitives for Computer Graphics
▶️ asia.siggraph.org/2023/presentat… Thurs 14 Dec @ 9am C4.6, Level 4
▶️ More info: gpu-primitives-course.github.io
2
51
138
💎 We'd also love to add your contributions to this blog to make it even more valuable 💎
Feel free to DM us, or reply to this post with your top tip for #graphics beginners.
If it's accepted, we'll publish it on the blog and credit your X handle (after confirming via DM).
Feel free to DM us, or reply to this post with your top tip for #graphics beginners.
If it's accepted, we'll publish it on the blog and credit your X handle (after confirming via DM).
12
4
28
We will be hosting a course: "#GPU Programming Primitives for Computer Graphics" at #SIGGRAPHAsia2023.
The course introduces programming primitives for parallel algorithms in computer graphics. It covers theoretical foundations, practical examples, as well as real-world e details, please refer to gpu-primitives-course.github.io
Looking forward to seeing you in person at SIGGRAPH Asia 2023!
The course introduces programming primitives for parallel algorithms in computer graphics. It covers theoretical foundations, practical examples, as well as real-world e details, please refer to gpu-primitives-course.github.io
Looking forward to seeing you in person at SIGGRAPH Asia 2023!
2
46
165
🧵Paper 2/2 introduces a novel approach to rendering reflections in real time that combines the radiance probes of an existing diffuse global illumination framework (GI-1.0) and denoised ray-traced reflections.
#SiggraphAsia2023
gpuopen.com/download/publi…
#SiggraphAsia2023
gpuopen.com/download/publi…
0
16
110
Two new #research papers today!
🧵Paper 1/2 introduces an easy method for improving the quality of VNDF sampling for reflections.
This simple method limits the sampling range for a state-of-the-art VNDF sampling method that uses a spherical cap.
gpuopen.com/download/publi…
🧵Paper 1/2 introduces an easy method for improving the quality of VNDF sampling for reflections.
This simple method limits the sampling range for a state-of-the-art VNDF sampling method that uses a spherical cap.
gpuopen.com/download/publi…
![Comparison between the previous VNDF sampling method [DupuyandBenyoub2023] (a) and our bounded VNDF sampling (b) for microfacet-based reflective surfaces. Our method efficiently reduces noise compared to the previous method while increasing the depth of paths, especially for rough surfaces.](/pic/orig/media%2FF80LiAMXoAU3URp.jpg)
2
41
144
We've just made the slide deck available for "Efficient Spatial Resampling using the PDF Similarity"!
Take a look: (95MB, PDF) 🧵2.5/3
gpuopen.com/download/publi…
Don't forget, there's more of our #research published at gpuopen.com/learn/publicat…
Take a look: (95MB, PDF) 🧵2.5/3
gpuopen.com/download/publi…
Don't forget, there's more of our #research published at gpuopen.com/learn/publicat…
0
14
37
3. "Subspace Culling for Ray-Box Intersection" proposes a novel culling approach to reduce false-positive hits for the bounding box by embedding a binary voxel data structure to the volume.
Preprint: gpuopen.com/download/publi… #i3D2023
More papers at gpuopen.com/learn/publicat… 🧵3/3
Preprint: gpuopen.com/download/publi… #i3D2023
More papers at gpuopen.com/learn/publicat… 🧵3/3
0
13
44
2. "Efficient Spatial Resampling Using the PDF Similarity" introduces a new method of rejecting spatial reuse based on the similarity of PDF shapes for single-bounce path connections such as direct illumination.
Preprint:
gpuopen.com/download/publi… #I3D2023 🧵2/3
Preprint:
gpuopen.com/download/publi… #I3D2023 🧵2/3
2
14
56
We're presenting 3 papers at #I3D2023! A thread 👇
1: "Performance Comparison of BVHs for GPU #RayTracing" proposes a novel method using global optimization with simulated annealing, and combines it with the existing best-performing BVH builders.
jcgt.org/published/0011… 🧵1/3
1: "Performance Comparison of BVHs for GPU #RayTracing" proposes a novel method using global optimization with simulated annealing, and combines it with the existing best-performing BVH builders.
jcgt.org/published/0011… 🧵1/3
3
43
143
👉Our #GDC23 #AMD sponsored sessions are TODAY!👈
Hope to see many of you #gamedev there! 👋
Find out more on GPUOpen: gpuopen.com/events/gdc-202…
Hope to see many of you #gamedev there! 👋
Find out more on GPUOpen: gpuopen.com/events/gdc-202…
3
11
37
Make time 🕦 for our #GDC23 temporal #upscaling session, covering:
⏫ Core concepts
⏫ How we implemented and evolved #FSR2 to work well in real games
⏫ An early look at where #FSR is going next 🕵️
Pack your very best motion vectors and join us! 👇
schedule.gdconf.com/session/tempor…
⏫ Core concepts
⏫ How we implemented and evolved #FSR2 to work well in real games
⏫ An early look at where #FSR is going next 🕵️
Pack your very best motion vectors and join us! 👇
schedule.gdconf.com/session/tempor…
3
19
81
Did you see "Two-Level Radiance Caching for Fast and Scalable Real-Time GI in Games" at the #GDC23 Advanced Graphics Summit yesterday? 🎇
Here's a PREVIEW video of a pre-release version of our GI-1.0 solution, including a look at glossy reflections!
youtu.be/o5zjvSTHz5s
Here's a PREVIEW video of a pre-release version of our GI-1.0 solution, including a look at glossy reflections!
youtu.be/o5zjvSTHz5s
2
27
87
As our #FidelityFX #opensource technology collection grows, we wanted to make them easier to evaluate, understand, and integrate into your games ✨
So we've reworked our FidelityFX technologies into a simple and consistent SDK, previewing at #GDC23 👇
schedule.gdconf.com/session/the-fi…
So we've reworked our FidelityFX technologies into a simple and consistent SDK, previewing at #GDC23 👇
schedule.gdconf.com/session/the-fi…
4
18
76
Microsoft #DirectStorage lets your game reduce loading times and improve asset streaming performance by maxing out the perf of modern high-speed NVMe storage.
Gain valuable DirectStorage integration tips to improve the gaming experience at #GDC23 👇
schedule.gdconf.com/session/direct…
Gain valuable DirectStorage integration tips to improve the gaming experience at #GDC23 👇
schedule.gdconf.com/session/direct…
0
20
78
If ISAs, multi-threading, and cache topologies are what gets you out of bed in the morning, come to our @AMDRyzen processor software optimization #GDC23 session!
Get the inside track on tuning for Zen 4's magnificent microarchitectural machinery: 👇
schedule.gdconf.com/session/amd-ry…
Get the inside track on tuning for Zen 4's magnificent microarchitectural machinery: 👇
schedule.gdconf.com/session/amd-ry…
2
23
63
Sparse distance fields are an efficient representation of complex scene geometry. They can help accelerate techniques like real-time GI 🌅
Our #GDC23 session walks you through the details, with an early look at our NEW #FidelityFX Brixelizer library! 👇
schedule.gdconf.com/session/real-t…
Our #GDC23 session walks you through the details, with an early look at our NEW #FidelityFX Brixelizer library! 👇
schedule.gdconf.com/session/real-t…
1
26
115
Our fantastic #Radeon Developer Tool Suite helps you to analyze and optimize @DirectX12 and @VulkanAPI games on PC 🛠️
At #GDC23 we're showcasing the tools and sharing what's new, including our work with @LunarGInc for #DXR support in GFXReconstruct 👇
schedule.gdconf.com/session/optimi…
At #GDC23 we're showcasing the tools and sharing what's new, including our work with @LunarGInc for #DXR support in GFXReconstruct 👇
schedule.gdconf.com/session/optimi…
0
8
52
We're very excited to be part of the Advanced #Graphics Summit again at #GDC23 this year!
Our session will cover a state-of-the-art, fast + high quality real-time global illumination (GI) system, researched right here at #AMD 🌞
Session details: schedule.gdconf.com/session/advanc…
Our session will cover a state-of-the-art, fast + high quality real-time global illumination (GI) system, researched right here at #AMD 🌞
Session details: schedule.gdconf.com/session/advanc…
0
7
80
Don't miss us at @Official_GDC 2023! 👋
Look out for tweets promoting our seven exciting sessions over the next few weeks. You can read more about what we'll be talking about here:
gpuopen.com/gdc-2023-annou…
Look out for tweets promoting our seven exciting sessions over the next few weeks. You can read more about what we'll be talking about here:
gpuopen.com/gdc-2023-annou…
1
13
60
If you develop for @RADV or @Steamdeck then do not miss this video by @Venemo & Friedrich Vock from @valvesoftware , and #AMD's @Reg__ at @VulkanAPI's #Vulkanised 2023.
Learn how to use our #Radeon Developer Tools to optimize your #gamedev!
youtube.com/watch?v=If7Dot…
Learn how to use our #Radeon Developer Tools to optimize your #gamedev!
youtube.com/watch?v=If7Dot…
1
10
50
Read our #realtime #denoising research!
"Weighted À-Trous Linear Regression (WALR) for Real-Time Diffuse Indirect Lighting Denoising": a new regression-based method for denoising images from path-traced #globalillumination with few rays/pixel. #raytracing
gpuopen.com/download/publi…
"Weighted À-Trous Linear Regression (WALR) for Real-Time Diffuse Indirect Lighting Denoising": a new regression-based method for denoising images from path-traced #globalillumination with few rays/pixel. #raytracing
gpuopen.com/download/publi…
2
27
104
Learn how your GPU submission strategy can affect #performance and what a #gamedev can do to make a positive difference!
This detailed PDF [7MB] was originally presented at @RebootDevelop Blue 2022 by a man with a Justin Bieber poster on his office wall.
gpuopen.com/presentations/…
This detailed PDF [7MB] was originally presented at @RebootDevelop Blue 2022 by a man with a Justin Bieber poster on his office wall.
gpuopen.com/presentations/…
0
37
76
Here's yet more #raytracing research!
"Performance Comparison of Bounding Volume Hierarchies for GPU Ray Tracing" empirically compares the most popular methods of addressing BVHs in the context of #GPU ray tracing.
🔗jcgt.org/published/0011…
📄 github.com/meistdan/hippie (HIP src)
"Performance Comparison of Bounding Volume Hierarchies for GPU Ray Tracing" empirically compares the most popular methods of addressing BVHs in the context of #GPU ray tracing.
🔗jcgt.org/published/0011…
📄 github.com/meistdan/hippie (HIP src)
0
25
95
We've published our #realtime #globalillumination rendering technique "GI-1.0" which uses two caching schemes. We use the ray budget efficiently to avoid wasting any computation.
The scene we tested takes only 1.9-3.2ms 👀
Read it now: 👇
gpuopen.com/download/publi… #raytracing
The scene we tested takes only 1.9-3.2ms 👀
Read it now: 👇
gpuopen.com/download/publi… #raytracing
9
105
538
Looking for more publications from our research groups? Our ARR Group has a new tech report, explaining the architecture and implementation of geometry and texture streaming in @Radeon #ProRender.
Find it linked here, alongside others you may have missed:
gpuopen.com/advanced-rende…
Find it linked here, alongside others you may have missed:
gpuopen.com/advanced-rende…
0
13
33
New! Preprints released by one of our #research groups for #SIGGRAPH Asia 2022:
⭐️ Fused BVH to Ray Trace Level of Detail Meshes (Tech report)
⭐️ Combining GPU Tracing Methods within a Single Ray Query
⭐️ Progressive Material Caching
Links on #GPUOpen 👇
gpuopen.com/advanced-rende…
⭐️ Fused BVH to Ray Trace Level of Detail Meshes (Tech report)
⭐️ Combining GPU Tracing Methods within a Single Ray Query
⭐️ Progressive Material Caching
Links on #GPUOpen 👇
gpuopen.com/advanced-rende…
1
25
104
@rtarun0 @antonschrein @panoskarabelas @TigerGD That's fair. Fortunately, it is a very short walk from Sponza to perform our required penance:
0
0
1