Guillaume Boissรฉ's Avatar

Guillaume Boissรฉ

@gboisse.bsky.social

Lead graphics programmer @AMD, previously SIE R&D (PlayStation). Opinions are my own. He/they. ๐Ÿ”— gboisse.github.io

326 Followers  |  46 Following  |  45 Posts  |  Joined: 21.10.2024  |  2.4081

Latest posts by gboisse.bsky.social on Bluesky

Preview
Introducing AMD GI-1.2 with multibounce indirect real-time rendering - AMD GPUOpen AMD GI-1.2, our real-time Global Illumination solution, is available now as part of the AMD Capsaicin Framework v1.3.

Capsaicin is not just for tacos, it's also our developer rendering framework!

v1.3 is out now:

๐ŸŒถ๏ธ Added heat with GI-1.2, which includes multiple-bounce indirect lighting.
๐ŸŒถ๏ธ HDR support with spicier tone mapping.

And much, much more: gpuopen.com/learn/gi-1-2...

20.11.2025 20:46 โ€” ๐Ÿ‘ 34    ๐Ÿ” 8    ๐Ÿ’ฌ 0    ๐Ÿ“Œ 1

Indeed, thatโ€™s another way of achieving the same effect:p

11.11.2025 17:53 โ€” ๐Ÿ‘ 0    ๐Ÿ” 0    ๐Ÿ’ฌ 0    ๐Ÿ“Œ 0
Post image Post image Post image

The โ€œalbedo bitโ€ is related to glossy reflections. Hereโ€™s an example: 1st shot is the cache without jitter (exposing the tiling), 2nd shot is with jitter, and you can see how the details disappearโ€ฆ Finally, 3rd shot divides the radiance by the average cell albedo and re-multiply with per-ray albedo.

11.11.2025 10:25 โ€” ๐Ÿ‘ 0    ๐Ÿ” 0    ๐Ÿ’ฌ 1    ๐Ÿ“Œ 0

Indeed, thatโ€™s the reason for this random selection. I make lists of hit points for every cell, then pick one hit at random (simple random access) that will be different each time. The process is similar to how I created reservoir lists in this paper (see โ€œalgorithm 2โ€): gpuopen.com/download/SA2...

10.11.2025 12:05 โ€” ๐Ÿ‘ 3    ๐Ÿ” 0    ๐Ÿ’ฌ 1    ๐Ÿ“Œ 0

I also average the albedo per cell, so I can then divide the resulting radiance by it and re-combine with the per-ray albedo (to recover texture details).
2. For direct lighting, itโ€™s all shadow maps still (although pending to change ๐Ÿ™‚) and cosine weighted sampling for indirect.
3. Yes, all diffuse.

10.11.2025 12:05 โ€” ๐Ÿ‘ 2    ๐Ÿ” 0    ๐Ÿ’ฌ 1    ๐Ÿ“Œ 0
Post image

Thatโ€™s a great article, thanks for sharing! Iโ€™ll try to answer all your questions.
1. Yes, thatโ€™s definitely a hard one, balancing between stability and responsiveness is really difficult. Iโ€™ve ended up with tracking temporally smoothed delta of luminance values (you can tweak all values of course):

10.11.2025 12:05 โ€” ๐Ÿ‘ 0    ๐Ÿ” 0    ๐Ÿ’ฌ 2    ๐Ÿ“Œ 0
Post image

New blog post! Behind the scenes of some of the techniques involved in making our last PC demo ๐Ÿ’ซ gboisse.github.io/posts/this-i...

05.11.2025 21:15 โ€” ๐Ÿ‘ 111    ๐Ÿ” 39    ๐Ÿ’ฌ 1    ๐Ÿ“Œ 1

Iโ€™ll take this as confirmation that youโ€™re preparing a new demo:)

22.10.2025 19:28 โ€” ๐Ÿ‘ 1    ๐Ÿ” 0    ๐Ÿ’ฌ 0    ๐Ÿ“Œ 0
A comparison between AMD Neural Light Sampling and ReSTIR.

A comparison between AMD Neural Light Sampling and ReSTIR.

Get up to speed with our latest advanced graphics research:

๐Ÿ“ GATE (Geometry-Aware Trained Encoding) for neural rendering.
๐Ÿ“ Real-time procedural GPU tree generation.
๐Ÿ“ Neural Visibility Cache for real-time light sampling.

๐Ÿ“ฅ Download the papers: gpuopen.com/learn/public...

16.10.2025 19:56 โ€” ๐Ÿ‘ 36    ๐Ÿ” 10    ๐Ÿ’ฌ 0    ๐Ÿ“Œ 0

Iโ€™d say that to me what you do is inspiring. although it likely wonโ€™t inspire me to restore old things:p still, I guess it counts?

04.10.2025 14:07 โ€” ๐Ÿ‘ 1    ๐Ÿ” 0    ๐Ÿ’ฌ 0    ๐Ÿ“Œ 0
Video thumbnail
04.10.2025 13:05 โ€” ๐Ÿ‘ 0    ๐Ÿ” 0    ๐Ÿ’ฌ 1    ๐Ÿ“Œ 0

lol, I did some tests a while ago with storing all neighbouring cell indices but my experience was that the extra memory fetch wasnโ€™t worth it (slower or barely faster than resolving the neighbours in the 1st place). Are you doing this at tile level?

26.09.2025 16:42 โ€” ๐Ÿ‘ 0    ๐Ÿ” 0    ๐Ÿ’ฌ 1    ๐Ÿ“Œ 0

Ticket booked, looking forward to it!

28.08.2025 21:09 โ€” ๐Ÿ‘ 9    ๐Ÿ” 1    ๐Ÿ’ฌ 1    ๐Ÿ“Œ 0
Post image

Performance is quite decent even at the highest quality settings; 120us for specular (and ~20us for diffuse):

27.08.2025 17:42 โ€” ๐Ÿ‘ 2    ๐Ÿ” 0    ๐Ÿ’ฌ 0    ๐Ÿ“Œ 0
Video thumbnail

Been wanting to implement that fast GGX-based convolution for glossy reflections probes for quite a while and finally made some time for it:) research.activision.com/publications...

27.08.2025 17:42 โ€” ๐Ÿ‘ 33    ๐Ÿ” 5    ๐Ÿ’ฌ 1    ๐Ÿ“Œ 0

Great work Krzysztof, congrats!

26.08.2025 20:37 โ€” ๐Ÿ‘ 1    ๐Ÿ” 0    ๐Ÿ’ฌ 0    ๐Ÿ“Œ 0
SIGGRAPH 2025 Course: Physically Based Shading in Theory and Practice

Most of the slides and course notes for the SIGGRAPH 2025 Physically Based Shading in Theory and practice course are now online, with more to follow:
blog.selfshadow.com/publications...

10.08.2025 14:46 โ€” ๐Ÿ‘ 111    ๐Ÿ” 55    ๐Ÿ’ฌ 1    ๐Ÿ“Œ 0

Great work, congrats!

28.07.2025 15:12 โ€” ๐Ÿ‘ 1    ๐Ÿ” 0    ๐Ÿ’ฌ 1    ๐Ÿ“Œ 0

I am! Userโ€™s guillaume1816 it seems.

22.07.2025 09:58 โ€” ๐Ÿ‘ 1    ๐Ÿ” 0    ๐Ÿ’ฌ 0    ๐Ÿ“Œ 0

Donโ€™t have any (hands-on) experience with surfels but definitely quite a bit with building real-time GI systems, happy to try and answer some questions, feel free to DM:)

22.07.2025 07:56 โ€” ๐Ÿ‘ 0    ๐Ÿ” 0    ๐Ÿ’ฌ 1    ๐Ÿ“Œ 0
Preview
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 RDNAโ„ข 4 architecture GPUs.

๐Ÿ’ก Do you want to learn more about using the matrix cores of AMD RDNAโ„ข 4 architecture GPUs?

Our quick how-to guide shows you how to:

โœ… Use WMMA intrinsics on RDNA 4.
โœ… Implement a simple MLP.

๐Ÿ‘‰ Read the guide: gpuopen.com/learn/matrix...

11.07.2025 15:59 โ€” ๐Ÿ‘ 17    ๐Ÿ” 4    ๐Ÿ’ฌ 0    ๐Ÿ“Œ 0
Preview
Guillaume Boissรฉ on X: "Small breakdown of the photon mapping effect:" / X Small breakdown of the photon mapping effect:

ha, nice! At some point I added โ€œde-interleavedโ€ shading and wrote some small breakdown: x.com/guitio2002/s... By now Iโ€™ve mostly moved away from light tracing tho, my personal lighting tech is now more akin to path tracing using ReSTIR, radiance caching, etc. Looking forward to see your progress:)

28.06.2025 15:54 โ€” ๐Ÿ‘ 1    ๐Ÿ” 0    ๐Ÿ’ฌ 1    ๐Ÿ“Œ 0

100,000 iirc. It was drawing a reflective shadow map from the spotlight, tracing the ~100k rays and shading using a mix of 2D light culling and stochastic sampling, followed by some spatiotemporal denoising. It was my first presentation at a conference (DevConโ€™17 in San Diego), quite fond memories:)

28.06.2025 14:49 โ€” ๐Ÿ‘ 2    ๐Ÿ” 0    ๐Ÿ’ฌ 1    ๐Ÿ“Œ 0
Post image

I did this, combined with some efficient light culling, to get some real-time dynamic GI on PS4 some nearly 10 years ago now:)

28.06.2025 14:26 โ€” ๐Ÿ‘ 1    ๐Ÿ” 0    ๐Ÿ’ฌ 1    ๐Ÿ“Œ 0

Ego-driven development:) Believing that the authors are incompetent is obviously oversimplifying, but it does make one feel smart about oneself.

24.06.2025 19:14 โ€” ๐Ÿ‘ 3    ๐Ÿ” 0    ๐Ÿ’ฌ 0    ๐Ÿ“Œ 0

Thanks, very appreciated! ๐Ÿ™

23.05.2025 20:16 โ€” ๐Ÿ‘ 0    ๐Ÿ” 0    ๐Ÿ’ฌ 0    ๐Ÿ“Œ 0

Good luck!

23.05.2025 19:52 โ€” ๐Ÿ‘ 0    ๐Ÿ” 0    ๐Ÿ’ฌ 0    ๐Ÿ“Œ 0
Visual node graph with ImGui | Guillaume's graphics blog Using Dear ImGui for creating a visual node graph.

Thanks! Itโ€™s a homemade graphics engine, mainly used for demoscene productions atm:) You can read more about it here if interested: gboisse.github.io/posts/node-g...

23.05.2025 19:02 โ€” ๐Ÿ‘ 6    ๐Ÿ” 2    ๐Ÿ’ฌ 2    ๐Ÿ“Œ 0
Post image

Quick behind the scenes of our latest PC demo production, hereโ€™s the node graph for the part from 02:14 to 03:00: youtu.be/W7Om9rf0qKc?... Might print out more graphs from other scenes if thereโ€™s some interest:)

23.05.2025 17:59 โ€” ๐Ÿ‘ 8    ๐Ÿ” 3    ๐Ÿ’ฌ 1    ๐Ÿ“Œ 0
Post image

Perks of the job:) #rdna4

22.05.2025 11:45 โ€” ๐Ÿ‘ 8    ๐Ÿ” 0    ๐Ÿ’ฌ 0    ๐Ÿ“Œ 0

@gboisse is following 20 prominent accounts