Darius Bouma

New blog post: A Decade of Slug This talks about the evolution of the Slug font rendering algorithm, and it includes an exciting announcement: The patent has been dedicated to the public domain. terathon.com/blog/decade-sl…

New blog post! Behind the scenes of some of the techniques involved in making our last PC demo 💫 gboisse.github.io/posts/this-is-…

Here is a new blog post of mine, where I dive deep into the theory and math behind ReSTIR GI. Big thanks to @nmkettunen for verifying the theory, reviewing the article, and catching a key detail I had missed. agraphicsguynotes.com/posts/understa…

.@TraverseBreda, one of the leading companies in graphics research and development, has unveiled Evolve, a groundbreaking GPU benchmark suite engineered to redefine how we measure and analyze graphics performance: 80.lv/articles/trave…

I just attended GPC. It was awesome, the talks were of high quality, the organization was great and I enjoyed hanging out with friends that I hadn’t see in a while. I can’t recommend it enough. If you are in Europe and enjoy graphics it’s a no brainer #contact" target="_blank" rel="nofollow noopener">graphicsprogrammingconference.nl/schedule/#cont…

We'll present "Hierarchical Light Sampling with Accurate Spherical Gaussian Lighting" at #SIGGRAPHASIA2024 #Tokyo gpuopen.com/download/publi… More info at shorturl.at/X0FcF

✨ We're very excited to be presenting at the first-ever Graphics Programming Conference #GPC2024 in Breda, NL, in November ✨ Find out more about our planned talks on GPU Work Graphs and Occupancy, plus other 🔥 sessions we'd like to highlight! gpuopen.com/gpc-2024/?utm_…

🎇 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…

Today mark the first day where we're publicly talking about our upcoming GPU benchmark suite called EVOLVE! We'll be at SIGGRAPH to spread the word, if you're interested in talking to us directly please book a spot: calendly.com/jasper-travers… Otherwise: evolvebenchmark.com

Brixelizer and Brixelizer GI are finally available to the public after almost two years of hard work 🎉

🧱 We are also introducing Brixelizer GI with the latest AMD FidelityFX SDK v1.1. 🧱 ✨Miss out on our #GDC2024 talk? You can read our new Brix GI blog now! gpuopen.com/learn/getting-… 🧵2/4

I'm excited to share our #SIGGRAPH2024 paper on unifying stochastic representations for light transport: cs.dartmouth.edu/~wjarosz/publi… We show how to use stochastic implicit surfaces to describe microfacet surfaces and participating media (and a bunch of new things in between).

Before implementing super cool Specular Polynomials, I started a small fun project and named it Stochastic Path Cuts😀 The ideas are taken from these inspiring papers: - Importance Sampling of Many Lights with Adaptive Tree Splitting fpsunflower.github.io/ckulla/data/ma… - Stochastic Lightcuts for Sampling Many Lights ieeexplore.ieee.org/document/91133… - Path Cuts: Efficient Rendering of Pure Specular Light Transport sites.cs.ucsb.edu/~lingqi/public…

Wow, this is so cool🙂 Specular Polynomials arxiv.org/pdf/2405.13409

@SebAaltonen Have you tried using interleaved gradient noise for your sampling pattern? This does a good job of representing the coverage in a local area of pixels.

My plan is to replace temporal white noise with even/odd frame noise. Fixed sampling pattern that provides maximum coverage over 2 frames. This reduces the maximum difference between frames and results in highest possible noise frequency, which is harder for our eyes to notice.

Last Friday I improved our SSAO spatial sampling patterns. Per pixel white noise for line directions and sample offsets causes lots of visible noise (left). Right image has 22.5 degree rotations with even/odd fixed 0.5 sample offset. 4x/2x less random. What about temporal?

I recently implemented both ReSTIR DI and ReSTIR GI as 2 separate pipelines. After some experimentation I figured I'd better fuse these two pipelines to fit my tight performance budget. The key to make this work was ...

After reading through the excellent ZH3 paper by @torust/@miwanicki/@Peter_shirley (torust.me/ZH3.pdf), I added the ZH3 hallucination method for calculating irradiance to both BakingLab and Probulator github.com/TheRealMJP/Bak… github.com/kayru/Probulat…

Made a first post in my personal blog! The topic is vertex quantization system in my engine daniilvinn.github.io/2024/05/04/omn…

We present A Fresnel Model for Coated Materials at Eurographics this year, where we propose an analytical RGB model for rendering coated conductors. Our method provides improved accuracy of Fresnel reflectance in BRDFs. I'll be presenting this paper Wednesday at 9:00 in Megaron A

@VrKomarov Looking nice, how well does it respond to changes in lighting conditions?

Integrated new Nvidia Sharc cache into my path tracer. Works pretty well. Need to make a few tweaks for volumetrics but it's almost working